u.s. doe arpa-e electricity research programs · hvdc links transformer based hvdc line lcc/vsc...

ARPA-E Electricity Research Programs

Tim HeidelProgram Director

Advanced Research Projects Agency – Energy (ARPA-E)

U.S. Department of Energy

PSERC Summer Workshop

July 15, 2015

The ARPA-E Mission

1

Ensure America’s

• National Security

• Economic Security

• Energy Security

• Technological Competiveness

Catalyze and support the development of

transformational, disruptive energy technologies

Reduce Imports

Reduce Emissions

Improve Efficiency

GENI

REACT

BEETIT

METALS

AMPED

GRIDS

HEATS

Solar ADEPT

FOCUS

ARID

MOSAIC

GENSETS

REBELS

ADEPT

SWITCHES

A portfolio enabling a secure, affordable, and sustainable electricity future.

NODES

GRID DATA

3

U.S. Carbon Emissions (2013): 5,390 MMT

4

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Wind

Solar/PV

Geothermal

Nuclear

Hydro

Natural Gas

Petroleum

Coal

Biomass

Biomass

Coal

Oil

Nat. Gas

Source: EIA

U.S. Primary Energy Consumption

5

0

20

40

60

80

100

120

Wind

Solar/PV

Geothermal

Nuclear

Hydro

Natural Gas

Petroleum

Coal

Biomass

U.S. Primary Energy Consumption (Quads)

July 16, 2015 6

U.S. Electricity Consumption

0

500000

1000000

1500000

2000000

2500000

3000000

3500000

4000000

4500000

1949

1953

19

57

1961

1965

1969

1973

19

77

1981

1985

1989

1993

19

97

2001

2005

2009

2013

Solar/PV

Wind

Geothermal

Waste

Wood

Nuclear

Petroleum

Natural Gas

Conventional Hydro

Coal

7Source: National Solar Radiation Database, NREL

Variability & Uncertainty

Emerging Power System Challenges

8

– Increasing wind and solar

generation

– Decentralization of generation

– Aging infrastructure

– Changing demand profiles

– Increasing natural gas generation

– Cybersecurity threats

‣ All of these challenges require

greater power system flexibility.

Evolution of Grid Requirements

GRID

Modernization

Affordable

Safe

Accessible

Reliable

Clean

Secure

Resilient

Flexible

Federal Power Act

1930s

Blackouts

1960s

Oil Embargo,

Environmental concerns

1970s

9/11

Stuxnet

2000s

Hurricanes

Katrina, Sandy

Polar Vortex

2010s

Increasing Dynamics

and Uncertainty

2020s

Source: Adapted from DOE Grid Tech Team

10

Responsive Demands

- Scheduling large loads (eg. industrial loads)

- Mobilize large numbers of small assets

Power Flow Controllers

- AC Power Flow Controllers

- High Voltage DC Systems

Energy Storage Optimization

- Scheduling energy flows

- Coordination of diverse storage assets

Transmission Topology Optimization

- Optimal line switching

- Corrective switching actions

‣ Advances in power electronics, computational technologies, and

mathematics offer new opportunities for optimizing grid power flows.

New Potential Sources of Network Flexibility

Project Categories

GoalsDuration 2012-2015

Projects 15

Total

Investment$39 Million

GENI ProgramGreen Electricity Network Integration

• Enable 40% variable generation penetration

• > 10x reduction in power flow control hardware (target < $0.04/W)

• > 4x reduction in HVDC terminal/line cost relative to state-of-the-art

• Power Flow Controllers

• Power flow controllers for meshed AC grids.

• Multi-terminal HVDC network technologies.

• Grid Optimization

• Optimization of power grid operation; incorporation of uncertainty into operations; distributed control and increasing customer optimization.

12

Power Flow Control White Space

• High costs and reliability problems are often cited against the widespread

installation and use of power flow control devices.

• New hardware innovations that can substantially reduce the cost of

power flow control devices are needed

• Fractionally rated converters (limited power device ratings).

Modular designs (increases manufacturability).

Series connected equipment with fail normal designs (gradual degradation).

Historically, power flow control devices have typically been manually

dispatched to correct local problems.

• New software advances that exploit new developments in

optimization and computational technologies are needed to enable

the real time coordinated, optimized dispatch of many power flow

control devices.

Power Flow Controllers

13

Inverter

Based Variable

Series Voltage

Injection

Power

Flow

Control

Toolkit

AC

Embedded

HVDC Links

Transformer

Based

HVDC Line

LCC/VSC

HVDC

Back-to-

Back

Multi-Terminal

HVDC

Variable Series

Reactance

Insertion

Xc

XL

Topology Control

(Line Switching)

Advances in Power Electronics (Components and Circuits)*

Fast, Reliable, Ubiquitous Communications

* ARPA-E’s ADEPT, Solar ADEPT, and SWITCHES programs

are funding some of these advances.

Grid Optimization Whitespace

‣ Existing grid optimization tools do not explicitly account for

variability and uncertainty

‣ Emerging need/opportunity to coordinate larger numbers of

distributed resources

‣ Recent advances enable more robust, reliable optimization of the

electricity grid despite:

• The limits of state estimation (finite numbers of sensors, limited models)

• Incomplete and imperfect information flow (due to latency of communications)

• Constrained computational resources for dynamic decision support (power

flow optimization is mathematically (NP) hard)

• Inherent uncertainties in price/market mediated transactions

• Physical constraints to control (cost, performance, and reliability of power

electronics).

14

Grid Optimization Opportunities

15

AC-Optimal

Power Flow

Stochastic

Optimization

Distributed

Optimization

Forecasting &

Dispatch of

Demand

Transmission

Switching

Energy

Storage

Optimization

Fast dynamic

stability

calculations

Grid

Optimization

Toolkit

Advanced Computing Architectures & Cloud Computing Infrastructure

New Optimization Methods & Advanced Solvers

GENI Portfolio

16

HVDC

EPR Cable

HVDC Multi-terminal

Network Converters

DC Breaker

HVDC Hardware

Power Flow Control Hardware

Cloud Computing

& Big Data

Topology Control

(Line Switching)

Distributed Series

Reactor

Magnetic Amplifier

Economic Optimization Contingency Management

Demand Response

Resilient Cloud /

Data Replication

Compact Dynamic

Phase Angle Regulator

Transformerless UPFC

Power System Optimization

Stochastic

Unit Commitment

Distributed Grid

Optimization (Prosumers)

Energy Storage

OptimizationAC-OPF

GENI Program Participants

17

Texas Engineering

Experiment Station

Lead Organizations

Subs/Team Members

CRA

GE

CIT

Varentec

Smart

Wire

AutoGrid

Univ of CA, Berkeley

Arizona

State Univ

Grid Protection

Alliance

Applied

Communication

Sciences

PJM

Boston UnivNortheastern

Tufts

Polaris Systems

OptimizationParagon Decision Technology

NCSU

SCE

Waukesha

Electric

Systems

MSU

BoeingInnoventor

New Potato

Technologies

Columbia Univ

Georgia

Institute of

Tech

RPI

MSU

LBNL

LLNL

ORNL

SNLGeneral

Atomics

CMU

Cornell

University of

Washington

Iowa St

Alstom

EPRITVA

Univ of CA,

Davis

WSU

UT-

Knoxville

ISO NE

OSI Soft

18

‣ Functions as a current limiter

to divert current from

overloaded lines to

underutilized ones

‣ Increases line impedance on

demand by injecting the

magnetizing inductance of a

Single-Turn Transformer

Distributed Series Reactors

‣ Two modes of operation:

– Autonomous (set point)

operation

– Two way communications

enabled for greater control

and line monitoring

19

TVA DSR Array Installation: October 15-20, 2012 99 Units

Field Testing Ongoing

Compact Dynamic Phase Angle Regulators

20

‣ Fractionally-rated converters (AC switches/ LC

filters).

‣ ‘Fail-normal’ bypass switch preserves system

reliability.

‣ 3-phase CD-PAR operation verified at 13 kV 1

MVA.

‣ Field test at Southern Company this year

‣ Target: $20-30/kVA of power controlled

‣ Modeling dynamic and steady-state impact of

CD-PAR at both distribution and transmission

Two-way wireless communication

Ain

Bin

Cin

Aout

Bout

Cout

+ +

Controller

Bypass Switch

13 kV isolated power supply

Gate drivers

Gateway PC

Fractionally rated power converter

13 kV Auto-transformer

Isolation Switch

Isolation Switch

No-bulk DC storage

21

Medium Voltage Mesh Power Flow Test Bed

5-Bus,12 kV meshed power flow test

bed built at NEETRAC .

22

Prototype (1-Phase) 7.5 MVA CVSR

1

1.5

2

2.5

3

3.5

4

4.5

5

5.5

0 200 400 600

AC

Win

din

g R

eac

tan

ce (

Ω)

DC “Control” Current (A)

Factory Test Results

ac 1500A

ac 1000A

ac 750A

ac 500A

ac 250A

Magnetic Amplifier for

Power Flow Control

• Inserts a controlled variable inductance

• Power electronics isolated from the HV

line

• Low power converter controls the high

voltage ac inductance

• Standard transformer manufacturing

methods

Transformer-less Unified Power Flow Controller

23

‣ Cascaded multi-level inverters (CMIs)

to eliminate transformers

– Modular, scalable design

– Low cost ($0.04/VA)

– Lightweight (1000 lbs/MVA)

– High efficiency (>99%)

– Fast dynamic response (<5 ms)

‣ Project goals:

Assemble and test 100 CMI sub-

modules

Prototype a 2-MVA CMI UPFC

Pilot test/demonstration of the

proposed UPFC on MSU’s

campus.

Transformer-less Unified Power Flow Controller

24

New Power System Programs in 2015

‣ NODES (Network Optimized Distributed Energy Systems)

– Reliably manage dynamic changes in the grid by leveraging flexible

load and Distributed Energy Resources’ (DERs) capability to provide

ancillary services to the electric grid at different time scales.

‣ GRID DATA (Generating Realistic Information for

Development of Distribution And Transmission Algorithms)

– Development of large-scale, realistic, validated, and open-access

electric power system network models (transmission and distribution)

that have the detail required for the successful development and

testing of transformational power system optimization and control

algorithms.

‣ OPEN 2015

– Transformational research in all areas of energy R&D, covering

transportation and stationary applications.

25

What happens after ARPA-E?

26

Energy Technology “Valleys of Death”

R&D Prototype Demonstration Commercialization

Investm

ent

Time

Valley of

Death #1:

Public

Funds

Valley of

Death #3:

Venture

Capital/

Private

Equity

Valley of

Death #2:

Follow-on

Development

Funds

27

Currently Seeking New Program Directors

ARPA-E is continually recruiting new Program Directors, who serve 3-year terms

‣ R&D experience; intellectual integrity, flexibility, and courage; technical breadth; commitment to energy; communication skills; leadership; and team management

‣ A passion to change our energy future

Please contact me If you are interested.

ROLES & RESPONSIBILITIES

Program development

‣ Perform technical deep dive soliciting input from multiple stakeholders in the R&D community

‣ Present & defend program concept in climate of constructive criticism

Active project management

‣ Actively manage portfolio projects from merit reviews through project completion

‣ Extensive “hands-on” work with awardees

Thought leadership

‣ Represent ARPA-E as a thought leader in the program area

ATTRIBUTES

29

www.arpa-e.energy.gov

Tim Heidel

Program Director

Advanced Research Projects Agency – Energy (ARPA-E)

U.S. Department of Energy

timothy.heidel@hq.doe.gov

Highly Dispatchable, Distributed Demand Response

Objectives: • Develop scalable SaaS platform

(DROMS-RT) for optimization and

management of Demand Response

(DR) programs.

• Standard communications protocols

for real-time DR signaling.

• Utilize machine learning to forecast

and characterize customer response

to DR signals.

• Optimization of economic dispatch

with DR.

30

Potential Impact: • Enable Real-Time predictive analytics on data from grid customers & sensors.

• Provide 90% reduction in the cost of operating DR programs by eliminating up-

front IT infrastructure costs for utilities.

• Increase the overall efficiency of DR operations by 30-50%.

Project Accomplishments

‣ Developed scalable cloud-based

platform for managing and

analyzing big-data sets for

energy applications.

‣ Demonstrated highly parallel

machine learning engine capable

of processing >1M forecasts on a

rolling basis every 10 min.

‣ Deployed commercial DR

programs up to 70K customers.

31

Post ARPA-E Goal: “Energy Data Platform”: 10X Reduction in the Cost

and Time to Market of Developing New IT/OT Apps

Transmission Topology Control Algorithms (TCA)

Objectives:

• Develop fast optimization, contingency analysis

and stability evaluation algorithms to allow grid

operators to optimize transmission topology (line

switching) in real time.

Potential Impact:

• Significantly lower generation costs

• Provide additional controls to manage congestion.

• Enable higher levels of renewable generation.

Accomplishments:

• Developed detailed model of PJM historical

operating hours (data supplied by PJM)

• Developed tractable topology control algorithms

using sensitivity-based heuristics and a reduced

MIP formulation of the TC problem. Practical

solution times reached (below 5 minutes in PJM).

• Developed fast state estimation and transient

stability algorithms to support the TC decisions

Estimated TCA Benefits

‣ Hybrid historical-high renewables simulation cases for PJM

– One power flow snapshot per hour for three representative historical weeks in 2010 (one per season)

– Renewables data from PJM Renewable Integration Study: 30% Low Off-shore Best sites On-shore Scenario

‣ Renewable curtailments were reduced on average by 40%

33

31%

64%

84%

98%

99%

Renewable curtailment reductions with Topology

Control Algorithms (Winter Week)

$0M

$2M

$4M

$6M

$8M

Summer Winter Shoulder

Production Cost Savings

Remaining Cost of Congestion

Weekly Cost of Congestion and Savings‣ Production cost savings in PJM RT markets under 2010 conditions were estimated to be > $100 million/year

‣ Savings of over 50% of Cost of Congestion

(Based on simulation results for three historical weeks)

Production Cost Savings = production cost without TCA (full topology)

– production costs with TCA

Cost of Congestion = production cost with transmission constraints

– production costs without transmission constraints