Identification and Congestion Analysis of Transmission Corridors of the Eastern

Interconnection

Aleksandr Rudkevich and

Minghai Liu

Presented at the 27th USAEE/IAEE Annual North American ConferenceHouston, TX

September 17, 2007

2

Background

• On August 8, 2006 the US DOE released National Electric Transmission Congestion Study

• Section 1221 of the 2005 Energy Policy Act requires DOE to conduct this study on a triennial basis

• On April 26, 2007, based on the results of the study, comments thereon and consideration of a broad set of economic, reliability and energy security criteria, the Secretary of Energy announced two draft designations of National Corridors: the Mid-Atlantic Area National Corridor and Southwest Area National Corridor

• The Mid-Atlantic Corridor encompasses large portions of the US power grid administered by two RTOs – the PJM Interconnection and New York ISO

• CRA International worked for DOE for more than a year to conduct an analysis of congestion in the Eastern Interconnection system to support that study and then to assist DOE with further analyses and with the response to industry’s comments

• In the process of conducting this work, we discovered that: – there is no adequate functional definition of a transmission corridor, – no analytical tools readily exist to analyze congestion of transmission corridors, – good and comprehensive data required to conduct the analysis are difficult to find, – the results of the analysis are voluminous and are difficult to summarize and present.

• In this presentation, we discuss the methodology and analyses conducted by CRA International on behalf of DOE underlying the designation of the Mid-Atlantic Area National Corridor

3

Transmission Congestion and its Implications• Congestion occurs whenever the provision of transmission

services required by the most economical generation/demand schedule exceeds the physical capability (thermal or voltage or stability) of the grid

• Congestion typically results in a redispatch of generation, preventing economically efficient power supply sources from serving consumer demand where it is needed

• Congestion leads to under-utilization of generation resources in constrained out areas

• Congestion results in high prices in constrained in areas

4

Congestion results in under-utilized generation: Difference in Capacity Factors PJM East vs. PJM West Source: CRA simulations

Figure 3. Capacity Factor by Resource Category: West PJM vs. East PJM

0.0%

10.0%

20.0%

30.0%

40.0%

50.0%

60.0%

70.0%

80.0%

90.0%

100.0%

$005 to$020

$020 to$030

$030 to$040

$040 to$050

$050 to$060

$060 to$080

$080 to$090

$090 to$100

$100 to$120

$120 to$150

Generation Cost Categories ($/MWh)

Cap

acity

Fac

tor

East PJMWest PJM

5

Congestion results in under-utilized generation: Difference in Capacity Factors NYISO Upstate East vs. Upstate West vs. Downstate Source: CRA simulations

Figure 7. NYISO: Capacity Factor by Region and by Cost Category (Using 2008 Base Case

Model Data)

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

$005 to$010

$010 to$030

$030 to$060

$060 to$070

$070 to$080

$080 to$090

$090 to$100

$100 to$120

$120 to$135

$135 to$200

$200 to$240

Cost Category ($/MWh)

DownstateUpstate EastUpstate West

6

Congestion Results in High Prices in Constrained in Areas: PJM LMPs by Zone

Round-the-Clock DA LMPs by Zone

0.00

20.00

40.00

60.00

80.00

100.00

120.00

1 2 3 4 5 6 7 8 9 10 11 12 1 2 3 4 5 6 7 8 9 10 11 12 1 2 3 4 5 6 7 8 9 10 11 12

2004 2005 2006

$/M

Wh

PEPCO

BGEDOM

AECOPSEG

RECO

DPLMETED

PECOPPL

JCPLAPS

PENELEC

AEPCOMED

DAY

DUQ

7

Congestion Results in High Prices in Constrained in Areas: NYISO LBMPs by Zone

NYISO RTC DAM LBMPs by Zone

0.00

20.00

40.00

60.00

80.00

100.00

120.00

140.00

160.00

1 2 3 4 5 6 7 8 9 10 11 12 1 2 3 4 5 6 7 8 9 10 11 12 1 2 3 4 5 6 7 8 9 10 11 12

2004 2005 2006

$/M

Wh

LONGILN.Y.C.DUNWODMILLWDHUD VLCAPITLMHK VLCENTRLNORTHGENESEWEST

8

Objectives of the Analysis

• Conduct a detailed simulation analysis of the Eastern Interconnection with the primary focus on the performance of the transmission system

• Identify major points of transmission congestion• Develop a practical definition of a transmission

corridor and use it to identify major corridors within Eastern Interconnection

• Develop quantitative indicators of congestion and relative importance for major corridors

• Identify significant corridors• Identify critical transmission constraints that limit

critical corridors

9

Analytical Steps

• Conduct an 8760-hour simulation of Eastern Interconnection using GE MAPS. Perform congestion analysis of transmission constraints/flowgates using simulation results

• Analysis was performed for two years, 2008 and 2011 and for three different fuel price scenarios

• Define key end markets (or hubs) for corridors, apply simulation results to hubs and compute major indicators for hubs

• Define corridors as pairs of connected hubs, screen for corridors of interest

• Establish a relationship between flowgates and corridors; measure corridor congestion; sort corridors using indicators of congestion and importance

• Identify critical constraints

11

Congestion Simulation and Analysis: Indicators of Congestion Applied to Flowgates/Constraints

• Binding Hours: Number of hours (or % of time annually) the flowgate is binding

• U90: Number of hours (or % of time annually) the flowgate is loaded in excess of 90% of its limit

• All-hours shadow price: average shadow price over all hours in a year

• Binding hours shadow price: average shadow price over hours during which the flowgate is binding

• Congestion rent: shadow price times flow summed over hours the flowgate is binding

• We selected 100 top flowgates in each category and merged them into 171 top flowgates of Eastern Interconnection

12

Defining hubs

• We envision transmission corridors as means of connecting end markets, or hubs, on the power grid

• Our approach in defining hubs varied by market:– For markets administered by NY ISO and ISO New England, we used their

respective LMP zones as a proxy for hubs. - This is because congestion typically occurs between these zones rather than

within these zones. - An exception is NYISO Zone J (New York City) but it has been decided not to

break Zone J into multiple hubs– For all other markets, we used FASTCLUS algorithm in SAS to cluster

generation and load buses into hubs according to their shift factors on all constraints/flowgates

– We sort these clusters by total weight, net generation weight, net load weight and select the top 90 bubs from each category

– In total we defined 255 hubs in Eastern Interconnection

13

Major Hubs in Eastern Interconnection

14

Top Constraints and Major Hubs

16

Transmission Corridors

• Why do we need to define transmission corridors?– “Fixing” a currently constrained element does not necessarily solve the problem, instead another

element may become constrained– A transmission corridor is a concept which helps to combine congestion problems into groups

such that when all problems within a group are solved, a significant congestion relief could be achieved in a system

• Geography is important to the extent the corridors connect geographically defined markets. Geography of corridors themselves (i.e. siting of transmission lines) was not addressed. Instead, corridors were defined electrically

• We did not focus on how to best solve congestion problems. We only focused on how to identify where the major problems exist now and how to assess/rank their significance among themselves

• There is no standard definition of a transmission corridor• Our approach is the following: for any two hubs find out if there is a corridor

connecting them and if so, which transmission facilities comprise that corridor• First we define a direct flow between two hubs A and B as a portion of the power

transfer from A to B that does not flow through any other hubs• Second, we postulate that a direct corridor between A and B exists if a direct flow

between A and B exceeds a certain threshold (we used 20%)

17

Definition: A direct corridor between two hubs exists if the direct flow from A to B exceeds 20%

HubA

HubB

Hub D

Hub C100%25%

40%

35%

In example below, 25% of transfer from A to B flows directly, the rest flows through other hubs

18

An Algorithm for Finding Corridors

• An algorithm for finding corridors has been developed and implemented in PowerWorld

• The algorithm identifies all direct corridors, computes direct transfers between the source and sink hubs and establishes for each corridor a list of transmission facilities that belong to the corridor

• We identified 3346 direct corridors in Eastern Interconnection• In addition, we considered composite corridors: paths comprised

of direct corridors such that the product of all direct flows along each path is in excess of 20%

• We found the total of 8004 composite corridors of which 3346 are direct

19

Significant corridors: filtering and classification rules

• Maximum net injection at the source hub should be above 1000 MW; Source hub must serve as a source for at least 4500 hours

• Maximum net withdrawal at the sink hub should be above 1000 MW; Sink hub must serve as a sink for at least 4500 hours

• Energy price at the sink should be at least $1/MWh higher than the average price at the source on average during the year

• 409 corridors “survived” these rules• All corridors originating at nodes containing under-utilized

generation are classified as source driven corridors• All corridors terminating at high price nodes are classified

as sink driven corridors

20

Designation and Analysis of the Draft Mid-Atlantic Corridor

• DOE combined results of the simulations analysis with historical analysis of congestion, review of regional planning documents and discussions with transmission planners

• (WECC conducted similar analyses of the Western Interconnection not discussed here)

• This process culminated in the release on August 8, 2006 of the National Electric Transmission Congestion Study

• Based on the industry comments with additional analytical support from CRA, on April 26, 2007 DOE designated two Draft Transmission Corridors:

– Mid-Atlantic Area National Corridor (includes some or all counties in DE, OH, MD, NJ, NY, PA, VA, WV, and DC); and

– Southwest Area National Corridor (seven counties in Southern California, three counties in western Arizona, and one county in southern Nevada)

• The rest of this presentation focuses on the Mid-Atlantic Corridor• At DOE’s request we further analyzed congestion in PJM and NYISO• First we identified hubs serving these markets which contain under-utilized

generation using simulation and historical data• Next we identified hubs experiencing high prices due to congestion• Using these subsets of hubs we identified all source-driven corridors and sink-

driven corridors

21

Geographical attributes and naming convention

• Geographical coordinates for generator and load buses at substation level were provided by the University of Illinois at Urbana-Champaign

• CRA then used a GIS database to identify for each bus its state, county and nearest metropolitan statistical area (MSA) as well as basic population and economic statistics

• GIS information was used to develop a naming convention of hubs:– hub name =

ACPF area name + unique cluster number +identifier such as

“MSA” if hub weight is predominantly in that MSA, or“County” if hub weight is predominantly in that county or“STA” if only the state name is apparent or“GEN” if cluster is named after the largest generating unit it contains

name of the identifier +“G” for clusters with higher generation weight and “L” for clusters with higher load weight

22

Example of PJM Source Driven CorridorsSource Source Node Sink Sink Node

PJM AEP_8_GEN_Amos01_G PJM AEP_1_MSA_Lynchburg_G

PJM AEP_8_GEN_Amos01_G PJM AEP_2_STA_TN-WV_L

PJM AEP_8_GEN_Amos01_G PJM AP_7_MSA_Hagerstown-Martinsburg_L

PJM AEP_8_GEN_Amos01_G PJM AP_8_MSA_DC-VA-MD_L

PJM AEP_8_GEN_Amos01_G PJM VAP_15_MSA_VB-Norfolk_L

PJM AEP_8_GEN_Amos01_G PJM VAP_27_MSA_DC_L

PJM AEP_8_GEN_Amos01_G PJM VAP_33_MSA_DC_L

PJM AP_1_County_Harrison_G PJM AP_7_MSA_Hagerstown-Martinsburg_L

PJM AP_1_County_Harrison_G PJM AP_8_MSA_DC-VA-MD_L

PJM AP_1_County_Harrison_G PJM PEPCO_1_MSA_DC_L

PJM AP_1_County_Harrison_G PJM VAP_15_MSA_VB-Norfolk_L

PJM AP_1_County_Harrison_G PJM VAP_27_MSA_DC_L

PJM AP_1_County_Harrison_G PJM VAP_33_MSA_DC_L

PJM DLCO_7_MSA_Pittsburgh_G PJM AEP_9_MSA_Canton-Massillon_L

PJM DLCO_7_MSA_Pittsburgh_G PJM AP_2_GEN_Albright3_L

PJM DLCO_7_MSA_Pittsburgh_G PJM AP_7_MSA_Hagerstown-Martinsburg_L

PJM DLCO_7_MSA_Pittsburgh_G PJM AP_8_MSA_DC-VA-MD_L

PJM DPL_8_GEN_Killen_G PJM AEP_1_MSA_Lynchburg_G

PJM DPL_8_GEN_Killen_G PJM AEP_2_STA_TN-WV_L

PJM EKPC_8_MSA_Maysville_G PJM AEP_2_STA_TN-WV_L

PJM NI_6_MSA_Chicago_G PJM AEP_1_MSA_Lynchburg_G

PJM NI_6_MSA_Chicago_G PJM AEP_2_STA_TN-WV_L

23

Example of PJM Sink Driven Corridors

Source Source Node Sink Sink Node

PJM PJM500_3_MSA_Pittsburgh_G PJM BGE_10_MSA_Baltimore-Towson_L

PJM PJM500_9_MSA_Pittsburgh_G PJM BGE_10_MSA_Baltimore-Towson_L

PJM PJM500_3_MSA_Pittsburgh_G PJM BGE_5_MSA_Baltimore-Towson_L

PJM PJM500_9_MSA_Pittsburgh_G PJM BGE_5_MSA_Baltimore-Towson_L

PJM PJM500_3_MSA_Pittsburgh_G PJM BGE_9_MSA_Baltimore-Towson_L

PJM PJM500_9_MSA_Pittsburgh_G PJM BGE_9_MSA_Baltimore-Towson_L

PJM PL_8_MSA_Allentown_G PJM JCPL_1_MSA_NewYork_L

PJM BGE_7_MSA_Baltimore-Towson_G PJM PECO_5_MSA_Philadelphia_L

PJM PJM500_4_MSA_Philadelphia_G PJM PECO_5_MSA_Philadelphia_L

PJM PJM500_7_MSA_York-Hanover_G PJM PECO_5_MSA_Philadelphia_L

PJM AP_1_County_Harrison_G PJM PEPCO_1_MSA_DC_L

PJM PJM500_3_MSA_Pittsburgh_G PJM PEPCO_1_MSA_DC_L

PJM PJM500_9_MSA_Pittsburgh_G PJM PEPCO_1_MSA_DC_L

PJM PENELEC_2_MSA_DuBois_G PJM PL_7_MSA_Allentown_L

PJM PJM500_3_MSA_Pittsburgh_G PJM PL_7_MSA_Allentown_L

PJM PJM500_9_MSA_Pittsburgh_G PJM PL_7_MSA_Allentown_L

NYPP NYISO_1_NYA PJM PSEG_3_MSA_NewYork_L

NYPP NYISO_3_NYC PJM PSEG_3_MSA_NewYork_L

PJM PJM500_4_MSA_Philadelphia_G PJM PSEG_3_MSA_NewYork_L

NYPP NYISO_1_NYA PJM PSEG_5_MSA_Philadelphia_L

NYPP NYISO_3_NYC PJM PSEG_5_MSA_Philadelphia_L

24

Example of NYISO Source Driven Corridors

Source Source Node Sink Sink Node

NYPP NYISO_1_NYA NYPP NYISO_11_NYK

NYPP NYISO_1_NYA NYPP NYISO_10_NYJ

NYPP NYISO_1_NYA NYPP NYISO_9_NYI

NYPP NYISO_1_NYA NYPP NYISO_7_NYG

NYPP NYISO_3_NYC NYPP NYISO_11_NYK

NYPP NYISO_3_NYC NYPP NYISO_10_NYJ

NYPP NYISO_3_NYC NYPP NYISO_9_NYI

NYPP NYISO_6_NYF NYPP NYISO_11_NYK

NYPP NYISO_6_NYF NYPP NYISO_10_NYJ

NYPP NYISO_6_NYF NYPP NYISO_9_NYI

NYPP NYISO_8_NYH NYPP NYISO_11_NYK

NYPP NYISO_3_NYC NYPP NYISO_7_NYG

NYPP NYISO_8_NYH NYPP NYISO_10_NYJ

NYPP NYISO_8_NYH NYPP NYISO_9_NYI

NYPP NYISO_6_NYF NYPP NYISO_7_NYG

25

Example of NYISO Sink Driven CorridorsSource Source Node Sink Sink Node

NEPOOL NEPOOL_2_NH NYPP NYISO_10_NYJ

NEPOOL NEPOOL_5_SEMA NYPP NYISO_10_NYJ

NEPOOL NEPOOL_7_CT NYPP NYISO_10_NYJ

NYPP NYISO_1_NYA NYPP NYISO_10_NYJ

NYPP NYISO_3_NYC NYPP NYISO_10_NYJ

NYPP NYISO_6_NYF NYPP NYISO_10_NYJ

NYPP NYISO_8_NYH NYPP NYISO_10_NYJ

ONTARIO IESO_15_G NYPP NYISO_10_NYJ

ONTARIO IESO_6_G NYPP NYISO_10_NYJ

ONTARIO IESO_7_G NYPP NYISO_10_NYJ

PJM PENELEC_2_MSA_DuBois_G NYPP NYISO_10_NYJ

NEPOOL NEPOOL_2_NH NYPP NYISO_11_NYK

NEPOOL NEPOOL_5_SEMA NYPP NYISO_11_NYK

NEPOOL NEPOOL_7_CT NYPP NYISO_11_NYK

NYPP NYISO_1_NYA NYPP NYISO_11_NYK

NYPP NYISO_3_NYC NYPP NYISO_11_NYK

NYPP NYISO_6_NYF NYPP NYISO_11_NYK

NYPP NYISO_8_NYH NYPP NYISO_11_NYK

ONTARIO IESO_15_G NYPP NYISO_11_NYK

ONTARIO IESO_6_G NYPP NYISO_11_NYK

ONTARIO IESO_7_G NYPP NYISO_11_NYK

PJM PENELEC_2_MSA_DuBois_G NYPP NYISO_11_NYK

26

Constraints Limiting Corridors• Let F1, F2, …, Fn be flowgates affecting a corridor

A B• The limit each flowgate Fj places on a corridor is

given by the formula:

• Using that formula we find the most limiting flowgate for the corridor A B in each hour

• Then we find the second and the third most limiting flowgate

• Among these # 1, #2 and #3 most limiting constraints we selected constraints that a) bind in the simulations and b) are limiting for at least 240 simulation hours (equivalent of 10 days)

• Constraints that are most limiting for source driven corridors are “source driven constraints”

• Constraints that are most limiting for sink driven corridors are “sink driven constraints”

• Many constraints are both source and sink driven

( | )

flowgate limit

flowgate flow

arg min index of the most

binding flowgate

j jj

j

j

j

jj

L FK

PTDF F A B

L

F

J K

−=

→

−

−

= −

27

Transmission Constraints Limiting the Ability to Use Generation in West PJM and East MISO and to Serve PJM Loads

ISOConstraint Number Constraint Cons type Class Congestion Symbol in Map

PJM 1 147 - Cloverdale-Lexington 5 Line Source & Sink —PJM 2 178 - Crete-E. Frankfort 345 Line Source —PJM 3 97 - Benton Harbor-Palisades Line Source —PJM 4 1-TRIPS,8MT STM -01PRNTY - 1 Line Source & Sink Top Constraint —PJM 5 78 - Black Oak-Bedington 500 Line Source & Sink Top Constraint —PJM 6 180 - Crete-St. John 345 B ( Line Source —PJM 7 314 - Homer City-Shelocta 23 Line Source & Sink —PJM 8 63601SOCIALBLAIRSVL Line Source & Sink —PJM 9 FG 1713 DICKERSN-PL VIEW 230 Line Source & Sink —PJM 10 460 - Mt. Storm-Doubs 500 (f Line Source & Sink —PJM 11 650 - Seneca-Maple 138 (flo) Line Source & Sink —PJM 12 188 - Danville-East Danville Line Sink —PJM 13 1-TRIP EDISON-MDWRD PBRG-TRN Line Sink —PJM 14 299 - Halifax-Person 230 (fl Line Sink —PJM 15 70 - Branchburg-Flagtown 230 Line Sink —PJM 16 751 - Warren-Falconer 115 (f Line Sink —PJM 17 Croyden-Burlington Line Sink —PJM 18 Mickleton-Delco Tap Line Sink —PJM 19 N PHILADELPH WANEETA ACTUAL Line Sink —PJM 20 130 - Cedar Grove-Clifton 23 Line Sink Top Constraint —PJM 21 Juniata-Lewiston Line Sink —PJM 22 NFG 23 - Roseland-Cedar Gro Line Sink Top Constraint —PJM 23 50 - Axton 765/138 Xfm (flo) Transformer Source & Sink

PJM 24 1-TRIPS,08SGROVE-08SGROVE- 1 Transformer Source

PJM 25 1130 - Wylie Ridge 345/500 X Transformer Source & Sink Top Constraint

PJM 26 317 - Homer City 345/230 Xfm Transformer Sink

PJM 27 690 - St. Clair 345/230 Xfm Transformer Sink

PJM 28 750 - Waldwick-Hawthorne 230 Transformer Sink

PJM 29 1228 - Clover 230/500 Xfm (f Transformer Sink

PJM 30 APS South Interface Interface Source & Sink Top Constraint —PJM 31 INTERFACE= PJM - CENTRAL Interface Source & Sink Top Constraint —PJM 32 INTERFACE= PJM - EASTERN Interface Source & Sink —PJM 33 INTERFACE= PJM - WESTERN Interface Source & Sink Top Constraint —PJM \ NYISO 34 FARRGUT 1000MW WHEEL Wheeling Source & SinkPJM \ NYISO 35 RAMAPO 1000MW WHEEL Wheeling Source & Sink Top Constraint

28

West PJM Source Driven Constraints (Locations of Under-utilized Generation in West PJM and Transmission Constraints leading to PJM Loads)

MISO

PJM

NYISO

Source & Sink

Sink

Source

Legend: ISO Boundary

Legend – Constraint Symbol

Legend: ACPF & Hub (PJM) Solid Pattern

Legend: ACPF & Hub (MISO) Hash Pattern

29

PJM Sink Driven Constraints (Transmission Constraints leading to PJM Loads)

Source and Sink driven constraints are consistent

MISO

PJM

NYISO

Source & Sink

Sink

Source

Legend: ISO Boundary

Legend – Constraint Symbol

30

Transmission Constraints Limiting the Ability to Use Generation in Upstate NYISO and Ontario and to Serve NYISO Loads

ISO

Constaint Number on Map Flowgate Name Constraint Type Class Congestion

Symbol in Map

NYISO 1 11 I/F UPNY - SENY OPEN LO Interface Source & Sink Top Constraint —NYISO 2 14 I/F WEST CENTRAL OP HI Interface Source & Sink Top Constraint —NYISO \ IMO 3 1454 - IMO-NYIS Interface Source & Sink —NYISO 4 1TRIP Dun-ShoreRd SpBrk-EGC Line Source & Sink Top Constraint —NYISO 5 1TRIP Leeds-Pleasant Val HI Line Source & Sink Top Constraint —NYISO 6 1-TRIPS,SCRIBA -VOLNEY - 1 Line Source & Sink —NYISO 7 1TSPBKTRMT:DUN SO1R-E179 ST Line Source & Sink —NYISO 8 7 I/F CENTRAL EAST LO Interface Source & Sink Top Constraint —NYISO 9 7 I/F MOSES SOUTH OPEN HI Interface Source & Sink Top Constraint —NYISO 10 8 I/F DYSINGER-EAST OPEN LO Interface Source & Sink —NYISO 11 Actual:E179 ST-HG 6 Line Source & Sink Top Constraint —NYISO 12 Actual:HUDAVE E-JAMAICA Line Source & Sink Top Constraint —NYISO 13 Actual:L SUCSPH-JAMAICA Line Source & Sink —NYISO 14 Actual:RAINEY8W-VERNON-W Line Source & Sink —NYISO 15 Actual:SPRBROOK-TREMONT Line Source & Sink —NYISO 16 Actual:V STRM P-JAMAICA Line Source & Sink Top Constraint —NYISO 17 CP10_12_1-tips, ReacBus-Dvnp Line Source & Sink Top Constraint —NYISO 18 CP10_20_E179St_Hg4_E179St_Hg Line Source & Sink Top Constraint —NYISO \ IMO 19 NFG7010 - IMO - ADIRONDACK Line Source & Sink Top Constraint —NYISO \ IMO 20 NFG7105 - ADIRONDACK - IMO Line Source & Sink Top Constraint —NYISO 21 NORHR138 138-NRTHPT P 138- 1 Line Source & Sink —NYISO 22 BARRET VALLEY STREAM ACTUAL Line Sink —NYISO 23 1-TRIPS,HMP HRBR-DVNPT NK- 1 Line Sink Top Constraint —NYISO 24 1-TRIPS,SHORE RD-L SUCS - 1 Line Sink —NYISO 25 Actual:DUN SO1R-E179 ST Line Sink Top Constraint —NYISO 26 VALLEY STRM - E GARDEN CTY Line Sink —NYISO \ ISONE 27 NFG9155 - NYIS-ISONE Interface Sink —NYISO 28 1TGOWNGOTN:GOWANUSS-GOTHLS S Line Sink —NYISO 29 1TGOWSGOTS:GOWANUSN-GOTHLS N Line Sink —NYISO 30 1-TRIPS,E15ST 46-FARRAGUT- 1 Line Sink —NYISO 31 16 I/F TOTAL EAST LO Interface Sink —NYISO 32 1-TRIPS,E VIEW1 -EASTVIEW- 1 Transformer Source & SinkNYISO \ PJM 33 FARRGUT 1000MW WHEEL Wheeling Source & Sink Top ConstraintNYISO \ PJM 34 RAMAPO 1000MW WHEEL Wheeling Source & Sink

31

NYISO Source and Sink Driven Corridors are also consistent - Locations of Under-utilized Generation in West and Upstate NYISO and Transmission Constraints leading to Downstate NYISO Loads

Shown in detail on next slide

32

NYISO Source and Sink Driven Corridors are also consistent - Transmission Constraints leading to Downstate NYISO Loads

33

Conclusions• Methodology described above is based on the objective, systematic and

mostly formalized analysis of transmission congestion over a very large system, Eastern Interconnection

• The “formalized” side of the analysis is both a blessing and a curse:– On one hand, formalization helps objectivity which is a key, given a highly politicized

environment in which transmission expansion decisions are being made– On the other hand, it is very difficult to include into a single formalized framework all

the complexities of the system spanning over multiple states and multiple markets• This study was a first in a series. The Energy Policy Act of 2005 requires

the DOE conducts a congestion study every three years. • We hope that the next study (due in 2009) will draw lessons from our

humble work, rely on the strong elements of our analytics and enhance not so strong sides of our study

34

Additional Information

• National Electric Transmission Congestion Studyhttp://nietc.anl.gov/

• Additional details on the analytical approach could be found in: “Identification and Congestion Analysis of Transmission Corridors of the Eastern Interconnection” by Aleksandr Rudkevich, Thomas Overbye, Kaan Egilmez, Minghai Liu, Prashant Murti, Poonsaeng Visudhiphan, and Richard Tabors, IEEE, Proceedings of the 40th Hawaii International Conference on System Sciences, January 2007