Pacific Northwest Smart Grid Demonstration Project Technical Status Update for GWAC Transactive Energy Workshop Ron Melton and Don Hammerstrom 2012.03.28 PNWD-SA-9681

1

Pacific Northwest Demonstration Project What:

• $178M, ARRA-funded, 5-year demonstration

• 60,000 metered customers in 5 states

Why: • Quantify costs and benefits • Develop communications protocol • Develop standards • Facilitate integration of wind

and other renewables

Who: Led by Battelle and partners including BPA, 11 utilities, 2 universities, and 5 vendors

2

Project Basics

Operational objectives Manage peak demand Facilitate renewable

resources Address constrained

resources Improve system

reliability and efficiency Select economical

resources (optimize the system)

Aggregation of Power and Signals Occurs Through a Hierarchy of Interfaces

3

Some Definitions

• Transactive Control A single, integrated, smart grid incentive signaling approach utilizing an economic signal as the primary basis for communicating the desire to change the operational state of responsive assets.

• Transactive Incentive Signal (TIS) A representation of the actual delivered cost of electric energy at a specific system location (e.g., at a transactive node). Includes both the current value and a forecast of future values.

• Transactive Feedback Signal (TFS) A representation of the net electric load at a specific system location (e.g., at a transactive node). Includes both the current value and a forecast of future values.

4

Role of a Transactive Node

• Respond to system conditions as represented by incoming Transactive Incentive Signals and Transactive Feedback Signals through

– Decisions about behavior of local assets – Incorporation of local asset status and other local information – Updating both transactive incentive and feedback signals

• Inputs are needed from node-owners to calculate incentive

and feedback signals • Each signal is a sequence of forecasts for a time-series, so

inputs will also be sequences of future (forecast/planned) values

5

End-to-End View of Transactive Control System – TIS Example Below is an example of a signal being modified as it flows from supply towards consumption through the transactive network

6

G

$ / M W h

Hours

$ / M W h

Hours

$ / M W h

Hours

Transmission - Level objectives

( e . g . Energy cost ; Trans Constraints )

Utility - Level objectives

( e . g . Avoid demand charges)

Local objectives ( e . g . Incent usage when local wind farm generating )

$ / M W

Hours

Final Incentive Signal received by Responsive Asset

Computing the Transactive Incentive Signal

(Energy Cost + Capacity Cost + Infrastructure Cost + Other Costs) TIS (t) = -------------------------------------------------------------------------------------- Total Energy Generated or Imported

7

Transactive Control Feedback Loop

New incentive signals and feedback signals are generated on an event-driven basis. The most recently available information is used. Each signal responds to changes in the other, and the values converge .

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

-24 -21 -18 -15 -12 -9 -6 -3 0

Load

or P

rice

(rat

io to

fina

l)

Time (hr from present)

LoadPrice

History of Load (TFS) & Price (TIS) Forecasts for Hour 0

8

Simple Example – Local EV Charging • Imagine the following situation:

– Three neighbors with electric vehicles and different charging strategies – All three fed by same distribution transformer – All three come home and want to do a fast charge at the same time!

• Problem – transformer is overloaded if all three fast charge at the same time

• Transactive control solution – – Transformer sees in feedback signal that all three plan to fast charge – Transformer raises value of incentive signal during planned charging

time to reflect decreased transformer life – Smart chargers and transformer “negotiate” through TIS and TFS until

an acceptable solution is found

9

$0.00

$0.01

$0.02

$0.03

$0.04

$0.05

$0.06

05

101520253035404550

16 17 18 19 20 21 22 23 24 1 2 3 4 5 6

Pric

e, T

IS ($

/kW

h)

Load

, TFS

(kW

)

Time of Day (hr)

EV3-Saver EV2-NowEV1-Flex Base LoadMax.Demand TFSCapacity TIS

Transactive Control – An Illustration

$0.00

$0.01

$0.02

$0.03

$0.04

$0.05

$0.06

05

101520253035404550

16 17 18 19 20 21 22 23 24 1 2 3 4 5 6

Pric

e, T

IS ($

/kW

h)

Load

, TFS

(kW

)

Time of Day (hr)

EV3-Saver EV2-NowEV1-Flex Base LoadMax.Demand TFSCapacity TIS

10

NW Region “Influence Map”--Topology

11

Cut Plane

Flowgate

Regional Modeling

Alstom EMS

Alstom MMS Future state Estimation by optimization

BPA

3TIER

Load Forecast Generation Schedules Outages

Network State

Gen. schedules Load forecasts

Transmission Zone TC Node Inputs

12

Project Nodes

Alstom Grid Models

Transmission Zone 5 Node

Portland General Node

Transmission Zone 14 Node

Lower Valley Node

Idaho Falls Node

Lower Valley Asset System(s)

Salem Site Asset Systems

Idaho Falls Asset System(s)

Regional Conditions

“Local Conditions” for Transmission Zone Nodes

TIS – down arrow TFS – up arrow

Asset system input – down arrow Local inputs – up arrow

13

Transactive Node Structure for Demo

TZ13- MT

ST10-Helena(BA09-NorthWestern

Energy)(UT09-NorthWestern

Energy)

ST11-Philipsburg(BA09-NorthWestern

Energy)(UT09-NorthWestern

Energy)

TZ12- Central Oregon TZ08-OR Cascades

TZ06-Northcentral Washington(BA04-BPA)

ST04-Ellensburg Renewable Park(Ut04-Ellensburg)

Canada@ Boundary

TS34 TS35

ST09-Milton-Freewater (UT08-Milton-Freewater)

TZ09- Southcentral OR

TS09

TZ04-Allston

TZ03-Paul

EasternMontana

TS03 TS06

FG02-N.Cascades North

FG01-Monroe-Echo Lake

FG04-R

aver Paul

FG05-PaulAllston

FG06-AllstonKeeler

FG10-W

. North of H

anford

FG08-W of Hatwai FG16-MT to NW

FG17-Lolo

FG20- LaG

rande

FG19-EnterpriseFG

18- W. of M

cNary

FG14-W. of Slatt

FG11-W

. of John Day

FG12-E. of John Day

FG07-S. Cascades

FG03-N.Cascades

South

FG09-E

. North of H

anford

Canada@ Custer

Wes

t CO

I

PD

CI

East COI

Wyoming

Nevada

FG15- Harney and Midpoint

Northern Montana

eg o a a d Subp oject a sact e Co t o odes & et o opo ogy

TS12

TS18 TS21 TS22

TS15

TS28 TS29

FG13

TS25

TIS/TFS Path (FGxx or TSxx)

TZ – Transmission ZoneBA – Balancing AuthorityUT – Utility (of Subproject)ST – Site (of Subproject)FG – Flowgate

Transactive Control (TC) NodeEIOC TC Nodes

ST02- UW Campus

ST01- Fox Island(UT01-Peninsula Light)

TZ02-West Washington(BA01-BPA) (BA02–SCL)(UT02-Seattle City Light)

TZ01-NW Washington

ST03-Salem(BA03-Portland General)

(UT03-PGE)

TZ05-Western OR

ST05-Reata(UT05-Benton PUD)

TZ07-Hanford(BA05-BPA)

TZ10-N. Idaho(BA07-BPA)

ST08-Haskil(UT07-

Flathead Electrid)

ST07-Libby(UT07-

Flathead Electric)

ST06-Pulman(BA06-Avista)(UT06-Avista)

ST14-DA &Energy

Management(UT11-Idaho Falls Power)

ST13- Loop Microgrid

(UT11-Idaho Falls Power)

TS33

ST12- Teton-Palisades

Power Interconnect(UT10-Lower

Valley)

TZ14- South Idaho(BA10-Pacificorp)

TZ11-NE Oregon(BA08-BPA)

14

Transactive Node Inputs & Outputs

15

Formalizing Transactive Control

• A formal model of transactive control has been designed with the following features:

– Scalable – Algorithmic – Support for interoperability

• A standardized approach is being promoted through design and implementation of a toolkit

– Well defined interfaces for utility asset systems – Simple, common, algorithms for updating transactive signals and

determining “control” signals to responsive asset systems

16

Bulk Power System Inputs to TIS Calculation 1.0 Imported electrical energy

1.1 Non-transactive imported energy

2.0 Renewable energy resource 2.1 Wind energy 2.2 Solar energy 2.3 Hydropower

3.0 Fossil generation 4.0 General infrastructure cost

5.0 System constraints 5.1 Transmission flowgate 5.2 Equipment and line constraints

6.0 System energy losses 6.1 Transmission losses 6.2 Distribution losses

7.0 Demand charges 7.1 BPA demand charges

8.0 Market impacts 8.1 Spot market impacts

17

Load Models

1.0 Bulk inelastic load 1.1 Bulk commercial load 1.2 Bulk industrial load 1.3 Bulk residential load 1.4 Small wind generator

negative load 1.5 small-scale distributed

generator negative load 1.6 Small-scale solar generator

negative load

2.0 General event-driven demand response

2.1 Commercial 2.2 Distribution system voltage

control 2.3 Residential behavior

2.3.1 Portals

18

3.0 General time-of-use demand response

3.1 Battery storage 3.2 Commercial 3.3 Residential behavioral

3.3.1 Portals 3.4 Residential 3.5 Distribution system voltage control

4.0 General real-time continuum demand response

4.1 Battery storage 4.2 Commercial 4.3 Residential behavioral

4.3.1 Portals 4.4 Residential

Load Models--continued

19

Cause and Effect Examples

• Three wind related scenarios:

– Case 1: Incentive for wind availability – Case 2: Incentive for wind ramp rate – Case 3: Incentive for balancing objective(s)

20

Case 1: Incentive for wind energy availability • Predicted incentive signal increases when wind energy

decreases and visa versa • Incentive is communicated and mixed between

transactive nodes • Assets respond to improve consumption of wind

– When wind energy is available – Near where wind is available

21

Transactive Node #1

Consider a Very Simple Topology

Transactive Node #2 22

Transactive Node #1

Assign Cost as a Function of Energy, Power and Time

Toolkit Resource Function #1

Toolkit Resource Function #2

P

$/M

Wh

P

$/M

Wh

Toolkit Resource Functions • Assign cost in a way that

will incentivize desired outcomes.

• Many different functions are possible, but acceptable functions must incur the same total cost over relatively long periods of time.

23

Power from these Resources is Predicted into the Future

Transactive Node #1

Toolkit Resource Function #1

Toolkit Resource Function #2

P

$/MWh

P

$/MWh

0

2

4

6

8

10

12

0 6 12 18 24 30 36 42 48 54 60 66 72

Pow

er (M

W)

Time (h)

Power Generated at Transactive Node #1

Wind Farm Base Generation24

Compare Two Ways to Assign Unit Cost to these Resources

0

2

4

6

8

10

12

Uni

t Cos

t ($/

MW

h)

Time

Unit Costs - Now

Wind Base Generation Aggregate

0

2

4

6

8

10

12

14

16

18

Uni

t Cos

t ($/

MW

h)

Time

Unit Costs - Transactive Control

Wind Base Generation Aggregate25

Compare How Costs Accumulate

0

20

40

60

80

100

120

Hour

ly C

ost (

$/h)

Time

Hourly Resource Costs

Wind - Now

Wind- Transactive Control

Base Generation

0

2000

4000

6000

8000

10000

12000

Cost

($)

Time

Cumulative Cost

Transactive Control Now26

These two options are

identical over time

The Weighted Incentive (TIS) follows the Energy Exported from this Location

TIS

Transactive Node #1

0

5

10

15

TIS

($/M

Wh)

Time

TIS

-2

3

8

13

18

Win

d Po

wer

(MW

)

Wind Farm

27

Responsive Assets and Toolkit Load Functions • Some system locations have controllable, responsive

generation and load assets • A “toolkit load function” is selected or created from scratch to

predict and model how the asset will respond as a function of – The incentive signal

• Absolute representation • Relative, statistical representation • History • Predictions

– Status of the asset – Other local information and conditions (e.g., weather)

28

To Neighboring Locations that have Demand-Responsive Assets

Transactive Node #2

Toolkit Load Function #1

Toolkit Load Function #2

Toolkit Load Function #3

Incentive - $/MWh

ΔP

0

Incentive - $/MWh

ΔP

0

Incentive - $/MWh

ΔP

0

P

Incentive - $/MWh

Bulk Inelastic Load Function #4

29

The Battery and Voltage Systems Figure Out When and How to Respond

• The battery and voltage control system used the entire predicted time horizon to determine when to best charge and discharge.

• The water heater system was not engaged by this modest event.

0

10

20

TIS

($/M

Wh)

TIS

TIS

-60

-40

-20

0

20

40

Chan

ge in

Pow

er (M

W)

Time

WH System Voltage Control Battery System

30

Case 2: Incentive for Wind Ramp Rate • Predict rate of change in predicted wind energy

– Function of first and/or second derivatives of predicted wind resource

• Increase incentive at times wind will be decreasing and visa versa

– Encourage other generation resources or even curtailment of load at times that wind energy is decreasing

– Discourage other generation resources or even increase load at times that wind energy is increasing

31

Case 2: Add Incentive for Rate of Wind Ramping

0

5

10

15

TIS

($/M

Wh)

Time

TIS

-50

-40

-30

-20

-10

0

10

20

30

40

Cha

nge

in P

ower

(MW

)

Time

WH System Voltage Control

Battery System

-2

3

8

13

18

Win

d Po

wer

(MW

)

Wind Farm

32

Applies where wind power changes

Increasingly richer

responses Superposed objectives and functions lead to …

Case 3: Incentive for Balancing Objective(s) • Create function to increment or decrement incentive

based on anticipated balance of resource and load – Increase incentive when there may be a deficit – Decrease incentive when there may be a surplus

• Address special circumstances, like times where wind farm production may become curtailed

33

Case 3: Add Incentives for Balancing Objectives

-100

-50

0

50

100

Bala

nce

Sign

al ($

/h)

Balancing Signal

Deficit of Power

0

5

10

15

20

TIS

($/M

Wh)

TIS

-75

-25

25

75

Chan

ge in

Pow

er (M

W)

Time (h)

WH System Voltage Control

Battery System

0

2

4

6

8

10

12

Win

d Po

wer

(MW

)

Time (h)

Wind Power with Balancing

Wind farm shares balancing responsibility

34

Surplus of Power

Signal modified to encourage others to share responsibility

Questions?

Ron Melton, Project Director ron.melton@battelle.org 509-372-6777 Don Hammerstrom, Principal Investigator donald.hammerstrom@battelle.org 509-372-4087 Project website: www.pnwsmartgrid.org

35