Eastern Wind Integration and

Transmission Study Overview

January 28th Webinar

Dave CorbusNational Wind Technology CenterNational Renewable Energy Lab

Eastern Wind Integration& Transmission Study Objectives• Evaluate the power system impacts and transmission

associated with increasing wind capacity to 20% and 30% of retail electric energy sales in the study area by 2024 ;

• Impacts include operating due to variability and uncertainty of wind; reliability

• Build upon prior wind integration studies and related technical work;

• Coordinate with JCSP and current regional power system study work;

• Produce meaningful, broadly supported results through a technically rigorous, inclusive study process.

Region Definition

Study area includes:

• PJM

• Midwest ISO

• Mid-Continent Area Power Pool

• Southwest Power Pool

• TVA

• New York ISO

• ISO New England

• Other interested parties

Project OrganizationDOE

SponsorNRELProject

Manager

Technical Review

Committee

Stakeholders

AWS ScientificWind

Modeling

EnerNex TeamEnerNex/MISO/VentyxProject Team - Analysis

Key Issues & Questions include

• What system operational impacts and costs are imposed by wind generation variability and uncertainty?

• What are the benefits from long distance transmission that accesses multiple wind resources that are geographically diverse?

• What are the benefits from long distance transmission that move large quantities of remote wind energy to urban markets?

• How do remote wind resources compare to local wind resources?

Key Issues & Questions include

• How much does geographical diversity help reduce system variability and uncertainty?

• What is the role and value of wind forecasting?

• What benefit does balancing area cooperation or consolidation bring to wind variability and uncertainty management?

• How does wind generation capacity value affect reliability?

Joint Coordinated System Plan (JCSP)

• The 2007/2008 Joint Coordinated System plan included MISO, PJM, SPP,TVA, MAPP, NYISO,ISO-NE plus other interested parties

• The JCSP performed a long term planning study incorporating both economic (2024) and reliability (2018) analysis of system performance for the combined JCSP areas

• The EWITS uses some of the model assumptions including the generation expansion (EGEAS modeling)

• Final summary report in progress

Wind Integration Methods & Best Practices

• Capture system characteristics and response through operational simulations and modeling;

• Capture wind deployment scenario geographic diversity through synchronized weather simulation;

• Match with actual historic utility load and load forecasts;

• Use actual large wind plant power statistical data for short-term regulation and ramping;

• Examine wind variation in combination with load variations;

• Utilize wind forecasting best practice and combine wind forecast errors with load forecast errors;

• Examine actual costs independent of tariff design structure.

• Examine impacts of BA consolidation and fast markets.

Key Tasks- Eastern Wind Integration & Transmission Study

• Mesoscale modeling

• Transmission Study

• Integration Study

Key Tasks - Mesoscale Modeling

– Identify wind sites

– Develop high quality wind resource data sets for the wind integration study area• Mesoscale modeling• 3 years of time series data (2004-2006)• 10-minute data at 2 km spatial resolution

– Develop wind power plant outputs

Mesoscale Grids

579 GWs of Wind Sites from Wind Site Selection process for EWITS

Offshore Wind

• Great resource• Well correlated with

load and close to load centers

• More expensive!

Deep Water Wind Turbine Development

Current Technology

Supply v. CF

Size Distribution• Maximum onshore plant sizes were

normally distributed between 100 MW and 1000 MW

• Additional “mega” sites (>1000MW)

0

50

100

150

200

250

300

100

200

300

400

500

600

700

800

900

1000

1100

1200

1300

1400

1500

Num

ber o

f Sit

es

Rated Capacity (MW)

Plant Size Distribution

Power Conversion TestingApproach

• Power conversion takes into account– Turbine power curve for site IEC class– Air density, turbulence– Wake and non-wake losses– Time filtering to replicate the “spatial smoothing”

of the output of a real wind plant

Plant Name StateRated Capacity

(MW)Turbine Type

Hub Height (m)

Blue Canyon I Oklahoma 74.25 NM72 (1.65MW) 67 m

Lake Benton Minnesota 103.5 Zond 750 51.2 m

Storm Lake I Iowa 112.5 Zond 750 63 m

Validation Sites

Validation Example Diurnal Patterns

00.050.1

0.150.2

0.250.3

0.350.4

0.450.5

0 5 10 15 20 25

Capa

city

Fac

tor

Local Time of Day

Blue Canyon I - 65M

MOD

OBS

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

0.45

0 5 10 15 20 25

Capa

city

Fac

tor

Time of Day

Storm Lake - 65M - Z750

MOD

OBS

Validation Example Mean Ramps

0.00

0.02

0.04

0.06

0.08

0.10

0.12

0.14

0 50 100 150 200

Mea

n A

bsol

ute

Dev

iati

on

Time Lapse (Minutes)

Storm Lake - 65M - Z750

MOD

OBS

0.00

0.02

0.04

0.06

0.08

0.10

0.12

0.14

0.16

0 50 100 150 200

Mea

n A

bsol

ute

Devi

ation

Time Lapse (Minutes)

Blue Canyon I - 65M

MOD

OBS

Mesoscale Output/Power Conversion

47.02225 -68.80990

DATE TIME 80M SPEED DIRECTION DENSITY TKE

20040101 0010 4.89790 270.09622 1.25625 0.03305

20040101 0020 4.94108 268.34360 1.25527 0.02336

20040101 0030 4.81025 267.33597 1.25509 0.01175

20040101 0040 4.89001 267.15210 1.25468 0.00649

20040101 0050 4.55865 265.43286 1.25415 0.00473

20040101 0100 4.70651 265.82401 1.25480 0.00252

20040101 0110 4.84289 269.14575 1.25461 0.00214

20040101 0120 4.85045 266.78668 1.25462 0.00247

20040101 0130 4.76209 266.21219 1.25440 0.00268

20040101 0140 4.74387 263.26474 1.25424 0.00220

20040101 0150 4.89790 260.24161 1.25384 0.00246

20040101 0200 4.93185 256.34119 1.25321 0.00351

20040101 0210 4.87496 252.86868 1.25324 0.00413

….

Why 20% and 30% Wind?

Regional Wind Requirements

0

10,000

20,000

30,000

40,000

50,000

60,000

70,000

80,000

90,000

100,000

(MW

)

reference 20% Wind 30% wind

reference 16,000 0 0 0 27,000 5,000 3,000 12,000 0

20% Wind 50,000 11,000 13,000 20,000 63,000 46,000 13,000 12,000 12,000

30% wind 75,000 18,000 20,000 30,000 95,000 68,000 19,000 18,000 18,000

MISO and

MROSPP

Entergy

TVA PJM SERC NYISO NEISO IESO

Scenario Development and Siting• Four Different Scenarios

– Three 20% and one 30% wind scenarios• Scenario that emphasizes development of local resources with

lower capacity factors• Scenario that emphasizes high capacity factor wind

development in the Midwest with larger transmission component

• All of four scenarios require a lot of wind and transmission! – Some offshore wind required

• NREL/AWS provided 700 GWs of wind plants and wind sites for the scenarios were picked from these– Scenario sites were selected from the “Superset” of 700

GWs of sites

Four EWITS Scenarios

• Scenario 1, 20% wind penetration – “Lowest Cost Wind”: Utilizes high quality wind resources in the Great Plans, with other development in the east where good wind resources exist. Total capacity in MISO, MAPP, and SPP is approximately 185 GW

• Scenario 2, 20% wind penetration – “Hybrid, with Offshore”: Some wind generation in the Great Plains is moved east, with capacity increased in PJM, NYISO, and ISO-NE. Some offshore development in the Northeast and Mid-Atlantic.

Four EWITS Scenarios

• Scenario 3, 20% wind penetration – “Load-weighted Wind Development, Aggressive Offshore”: More wind is moved east toward load centers, necessitating even more utilization of off-shore resources.

• Scenario 4, 30% wind penetration – “Aggressive On- and Off-shore”. Meeting the 30% energy penetration level uses a substantial amount of the higher quality wind resource. Lots of offshore is needed to reach the target energy level, and the capacity in MISO, MAPP, and SPP goes back up to 188 GW.

EWITS Scenario Installed Wind Capacity by Region

0

10000

20000

30000

40000

50000

60000

70000

80000

90000

100000

MISO/MAPP SPP TVA SERC PJM NYISO ISO-NE

Region

Inst

alle

d C

apac

ity

(MW

)

Scenario 1

Scenario 2

Scenario 3

Scenario 4

20% wind Scenarios

64,100230,29916,620225,902-223,609Total Capacity

11,04024,9275,00013,837-4,291ISO-NE

9,28023,1672,62016,507-7,742NYISO

3978078,7365,00033,192-22,669PJM

4,0005,0094,0005,009-1,009SERC

-1,247-1,247-1,247TVA

-39,328-69,804-66,576SPP

-57,885-86,306-120,075MISO/MAPP

Offshore (MW)Total (MW)Offshore

(MW)Total (MW)

Offshore (MW)

Total (MW)

3: Load-weighted Wind Development, Aggressive Offshore

2: Hybrid w/Offshore1: Lowest Cost Wind

Region

64,100230,29916,620225,902-223,609Total Capacity

11,04024,9275,00013,837-4,291ISO-NE

9,28023,1672,62016,507-7,742NYISO

3978078,7365,00033,192-22,669PJM

4,0005,0094,0005,009-1,009SERC

-1,247-1,247-1,247TVA

-39,328-69,804-66,576SPP

-57,885-86,306-120,075MISO/MAPP

Offshore (MW)Total (MW)Offshore

(MW)Total (MW)

Offshore (MW)

Total (MW)

3: Load-weighted Wind Development, Aggressive Offshore

2: Hybrid w/Offshore1: Lowest Cost Wind

Region

30% Wind Scenario

79,100337,708Total Capacity

11,04024,927ISO-NE

9,28023,167NYISO

54,78097,736PJM

4,0005,009SERC

-1,247TVA

-69,309SPP

-120,313MISO/MAPP

Offshore (MW)Total (MW)

4: Aggressive On- and Off-shore

Region

79,100337,708Total Capacity

11,04024,927ISO-NE

9,28023,167NYISO

54,78097,736PJM

4,0005,009SERC

-1,247TVA

-69,309SPP

-120,313MISO/MAPP

Offshore (MW)Total (MW)

4: Aggressive On- and Off-shore

Region

Scenarios and Siting Things to keep in mind during the study

• How much capacity can be reasonably exported (imported) at each area?

• Assume constant energy between scenarios rather than constant number of plants:– Typical Great Plains capacity factor/Ohio C.F.=

45/30 = 1.5– ~Roughly 2 MW of wind in Great Plains produces

the same energy as 3 MW of wind in Ohio.

EWITS Scenario 1 Generation Siting

EWITS Scenario 2 Generation Siting

EWITS Scenario 3 Generation Siting

EWITS Scenario 4 Generation Siting

Key Tasks- Develop Transmission Plan

– JCSP reference future and 20% wind and 30% wind scenarios

• Builds on JCSP work

– Analyze different transmission alternatives for different wind scenarios

• 765 AC and HVDC• High in-state wind

versus high wind exports

Develop Preliminary Transmission Plans

• Use JCSP 20% wind scenario transmission overlay as the starting point to develop initial plans for EWITS four scenarios

• Determine type, size and route of transmission lines

• Determine costs and land requirements

• Determine potential substation and DC terminal locations

Scenario 1 20% Lowest Cost Wind

Scenario 1 Constrained Case Annual Gen Weighted LMP

Scenario 1 Constrained Case Annual Load Weighted LMP

Scenario 1 Generation Difference: Copper Sheet Minus Constrained Cases

Scenario 1 Top 24 Interfaces with Largest Annual Energy Difference

Scenario1

AMRN - IN 252,464 -17 28,856IN - OH 185,865 0 20,843OH - EPJM 180,721 -12 18,662AMRN - IOWA 0 -131,654 15,173SPP - SPS 248 -91,858 13,482SPP - EES 104,579 -3 12,551TVA - EES 95 -81,698 10,173WAPA - MINN 67,140 -11 9,243ISONE - NYISO 368 -63,842 9,128SOUTHERN - TVA 1 -76,910 9,045ATC - AMRN 76,404 -1 8,771MINN - ATC 74,025 0 8,647AECI - SPP 13 -68,241 8,565NYISO - PJM 876 -60,431 8,457IESO - NYISO 43,378 -11 4,971MICH - IN 160 -38,925 4,937AMRN - AECI 386 -24,081 4,585AMRN - TVA 39,604 0 4,571SPP - AMRN 37,948 -3 4,556WAPA - IOWA 30,887 -6 4,343MINN - IOWA 29,318 -111 4,102MICH - IESO 27,567 -479 4,070IOWA - SPP 1,594 -20,964 3,978AECI - EES 31,911 0 3,607

Coppersheet Minus Constrained

INTERFACETotal Positive Energy (GWh)

Total Negative Energy (GWh)

Additional Transfer Needed to Deliver 80%

Energy (MW)

Joint Coordinated System Plan Overlay – 20% Wind Scenario

4. Wind Integration Study• Evaluate operating impacts

– Regulation– Load Following– Unit Commitment

• Evaluate reliability impacts (ELCC/LOLP)• EWITS is first and foremost a wind

integration study– What are the integration costs and issues for 20

and 30% wind?– How is other generation affected?

Production Simulation Methodology

Case comparison approach– Actual wind vs. “ideal” wind– Objective is to determine relative value of two

resources providing same amount of annual/daily energy

Issues– Approach is established as best way to

accomplish objective– Not been attempted on this scale before

Hourly Modeling Objective– Chronological simulation of operational planning and power

system operation– Mimic

• Day-ahead unit commitment and scheduling based on load and wind generation forecasts

• Real-time operation with actual wind and load How do we simulate the Eastern Interconnection in 2024?

– Period-ahead planning (e.g. day-ahead unit commitment)– Real-time operations (at minimum of hourly granularity)– Operational structures

• Conventional control areas?• Existing markets?

Hourly Modeling PROMOD capabilities

– Reserve modeling• Types• Treatment (e.g. variable by hour?)

– Commit based on forecast, simulation based on actual quantities?

– Features for treatment of uncertainty?

Modeling Transactions– Day-ahead and “real time”– Relevant program features

Intra-Hour Impacts Objective

– Determine operating reserves required to manage control area with wind generation

– Feed requirements forward into hourly modeling

Variability of wind generation adds to existing variability, increasing requirements for RT ancillary services

Analytical approach– Based on high-resolution (< 10 min) load and

wind generation data

LOLP and ELCC Analysis

Objective– Determine contribution of wind generation to Eastern

Interconnection reliability

– Assess reliability value of transmission only(?)

Issues– Transmission overlay could have significant impact on existing

LOLE zones

– Transmission will serve as capacity resources for some zones; may make some zones very reliable, such that ELCC of wind would be minimal

Predecessor tasks– Requires PROMOD to determine new area import limits

– GE MARS model to be developed from PowerBase

– Resource constraints may necessitate staging

Reliability Analysis GE MARS

– Monte-Carlo based chronological reliability simulation

– Now in use at MISO

Objectives– Calculate ELCC for wind generation based on

comparative LOLE cases– Zone-by-zone basis

Input data– Network, resource, and load data input developed

from PowerBase– Wind as load modifier

Downloading EWITS Wind Data http://wind.nrel.gov/public/EWITS/

Download Time-series DataAll data (using ftp site)EWITS ftp site (ftp://ftp2.nrel.gov/pub/ewits) Instructions for ftp  Data for individual sites using interactive websiteEWITS Interactive Website  Frequently-Asked Questions (FAQs) about time-series

dataEWITS FAQ

Downloading EWITS Wind Data EWITS Interactive Website

/

Eastern Wind Integration & Transmission Study

Schedule• Nov 07 – Feb 08 Study Development

• March 2008 Award Wind Mesoscale Modeling Contract

• July 2008 Award Wind Integration Contract

• April – Oct 2008 Develop Wind Data Sets

• Sept 08 – June 2009 Evaluate Operating & Reliability Impacts; Develop Transmission

Plan

• August 2009 Complete Study

Your Input is Important!

• EWITS Website - http://wind.nrel.gov/public/EWITS/

• Suggestions on questions to address in study

or other comments/input

• Contact Dave Corbus at David_Corbus@nrel.gov (303-

384-6966) or Matt Schuerger at

MattSchuerger@earthlink.net (651-699-4971)