WESTERN INTERCONNECTION

TRANSMISSION PATH FLOW STUDY

1998 thru 2005

September 2007

WECC – Transmission Expansion Planning Policy Committee (TEPPC)

Historical Analysis WG

2

TABLE OF CONTENTS

I. INTRODUCTION ------------------------------------------------------------------------------------ 3

II. PATH FLOW METHODOLOGY --------------------------------------------------------------- 4

a. Paths Analyzed -------------------------------------------------------------------------------- 4 b. Analysis Methodology and Indices ------------------------------------------------------- 5 c. Path OTC Assumptions --------------------------------------------------------------------- 6

III. PATH FLOW ANALYSIS AND RESULTS Summary Graphs – 75% and 90% Indices -------------------------------------------------- 10 Summary Maps ------------------------------------------------------------------------------------ 10 Individual Path Results – Seasonal 75% Indices 1998 thru 2005 ---------------------- 11 Individual Path Results – Path Flow Frequency Distributions 2004-05 --------------- 11

IV. TREND ANALYSIS AND RESULTS

1998 through 2005 -------------------------------------------------------------------------------- 17

V. OBSERVATIONS AND CONCLUSIONS -------------------------------------------------- 22 APPENDIX – included as a separate EXCEL file

A. Path Summary Charts – a. % Time Exceeding 75% of OTC b. % Time Exceeding 90% of OTC

B. Individual Path Results – Seasonal 75% Indices 1998 thru 2005 C. Individual Path Results – Path Flow Frequency Distribution 2004-05 D. Tabulated Analysis Results

3

SECTION I

INTRODUCTION

The analysis of actual historical power flow data for the transmission system in the Western Interconnection provides an indication of how marketers and load serving entities have utilized the transmission system to market energy and serve load. This information is also useful in the analysis and identification of potential future areas of congestion and for verifying model representation for power flow and production costing analysis. Although the flow is a useful indicator, this may not reflect the schedules on a path, and cannot reflect high value uses of transmission to allow access to operating reserves, forced outage reserves or market options. This work is the latest in a continuing effort to analyze utilization of the western Interconnection transmission system. The WECC Transmission Expansion Planning Policy Committee’s Historical Analysis WG performed the analysis in this report. Similar analysis was performed previously by the three western Regional Transmission Associations (RTA) as documented in the latest Biennial Transmission Plan, dated July 2000 and by the Seams Steering Group – Western Interconnection, in their report dated February 2003. The time period covered by the analysis in this report is from Winter 1998-1999 through Summer 2005 with emphasis on the 2004-05 time period. Results of this analysis were provided to the US Department of Energy by the Western Congestion Assessment TF in May 2006, to assist with the US DOE study of transmission congestion in the Western Interconnection and Eastern Interconnections. The analysis was performed for 28 transmission paths, representing most of the major transmission paths in the western interconnection. The analysis utilized real time hourly power flow and operating transfer capability data submitted by path operators and archived in the WECC EHV Data Pool database. Most data in the EHV Data Pool database is complete. In some cases, the real time path operating transfer capability was not reported and assumptions had to be made for this analysis, based upon published path transfer capabilities. The OTC assumptions for each path are noted in this report. Hourly path flow data for Canada to NW and the Pacific HVDC Intertie was also obtained from the Bonneville Power Administration, BC Hydro and the Alberta Electric System Operator (AESO). To facilitate comparison among the paths, a utilization indicator was calculated. The same indicator was used in the RTA’s Biennial Transmission Plan report. This indicator is calculated as the percentage of time the path exceeds 75% of its operating transfer capability. The 75% level was chosen as an indication of a path that may be considered heavily utilized. A 90% probability level was also calculated for the 2004-05 time period. These indices were developed for purposes of this analysis and have no basis in terms of an accepted industry standard or practice. These indices were used by the US DOE as a congestion measure in their August 2006 Congestion Report. The magnitude of the indicators is not necessarily an indication that there is congestion, or an inability to meet the needs of transmission users, on the path. In the western interconnection, paths are

4

designed to be loaded to 100% of their operating transfer capability and withstand a credible outage without violating reliability standards.

SECTION II

PATH FLOW METHODOLOGY

Actual archived MW power flow data for the major transmission paths in the Western Interconnection were analyzed from winter 1998-99 through summer 2005, using data from the Western Electricity Coordinating Council (WECC) EHV Data Pool database. Analysis of the data was performed in late 2005 and early 2006 and therefore does not include the analysis of 2006 or later data. Future updates will include an analysis of 2006 and 2007 data. Results are presented in this report by season and by individual path. Information from prior reports was used for analysis prior to April 2004. The data analyzed consisted only of actual hourly MW flows. Hourly ATC and schedule data was not available and therefore was not included in this analysis. Analysis of ATC and schedule data is required to assess path congestion and whether paths are heavily scheduled or whether contract rights are being scheduled. A path may experience low actual flows depending upon the magnitude of its hourly schedules; yet it may be congested to the market if the path is fully reserved on a firm basis (zero firm ATC). The purpose of the analysis was to determine the historical utilization of the major transmission paths in the Western Interconnection. It should not be concluded from this analysis that it is either necessary or economical to take any corrective actions for the loading levels reported. The results may, however, provide information for identifying paths for further study regarding the potential benefits and costs of increasing their path capacity. Actual flows were calculated on a per unit basis, referenced to the paths Operating Transfer Capability (OTC). Table I identifies the OTC used for each path. Hourly OTC data was used unless it was not available. Results are presented by seasons, defined as:

Winter – November thru March Spring – April and May Summer – June thru October

Paths Analyzed Data for the following paths were analyzed. Path names and path numbers are from the WECC Project Rating Catalog. A map showing the geographic location of the individual paths is included in Figure 2. A list of the lines making up each path may be found in the WECC Path Rating Catalog.

5

WECC Path # WECC Path Name

1 Alberta - BC 3 Northwest – Canada

8 Montana to Northwest 14 Idaho to Northwest 15 Midway – Los Banos 17 Borah West 18 Idaho – Montana 19 Bridger West 20 Path C 22 Southwest of Four Corners 26 Northern – Southern California 30 TOT 1A 31 TOT 2A 34 TOT 2B 35 TOT 2C 36 TOT 3 45 SDG&E - CFE

47 Southern New Mexico 48 Northern New Mexico

49 East of Colorado River (EOR) 50 Cholla – Pinnacle Peak 51 Southern Navajo 65 Pacific DC Intertie (PDCI)

66 COI Analysis Methodology and Indices The WSCC EHV Data Pool database was used for the frequency distribution analysis. The analysis period for the frequency distribution analysis was from April 2004 thru October 2005. Yearly trend analysis utilized data from the 2000 Biennial Transmission Plan, from the February 2003 report as well as updated information for the April 2004 through October 2005 time period.

Table I identifies the OTC assumption for each path.

Frequency Distribution The percentage of time a path exceeded a given percentage of its OTC was calculated and presented as a frequency distribution plot for each transmission path, using the hourly MW flow data in the EHV Data Pool database. Plots for the individual paths are presented in the Appendix. Percentage of time Exceeding 75% and 90% of OTC Beginning with the 2000 Biennial Transmission Plan, a utilization indicator was developed. It is defined as the percentage of time over the season that the path loading exceeds 75% of the path OTC. Based upon WECC reliability criteria, a path may be loaded up to its OTC level and be able to withstand various outage contingencies without violating reliability criteria. Experience has shown,

6

however, that loadings above approximately 75% of OTC may be associated with paths considered to be heavily utilized from a marketing or commercial use viewpoint. The 75% value is not based upon any industry standard or guide, and was chosen primarily to establish a figure of merit for ease of loading comparison between paths. The derivation of the “Percentage of time exceeding 75% of OTC” value is graphically illustrated in Figure 1. Previous reports also reported the peak loading for each path, however this was not done in this report. The “definition” of peak loading used in previous analysis is also shown in Figure 1.

OTC Assumptions Flow analysis is presented as a percentage or per unit of path Operating Transfer Capability (OTC). Hourly OTC values are reported in the EHV Data Pool database. Those reporting path OTC are supposed to calculate the OTC each hour, adjusting reported values for changing operating conditions. This is not always the case, however. In some cases, no OTC values are reported by the path operator and a zero value is included in the database. In some cases, an OTC value is reported each hour, however the value is the same for each hour and is not varied as operating conditions change. In some cases, the OTC data is reported correctly in which case the reported OTC value does vary in magnitude, indicating that the path operator is adjusting the path capability for changing operating conditions. Table I summarizes the OTC value used in this analysis. If a value is reported by the path operator, whether it is adjusted for changing operating conditions or not, this value is used. If no value is reported each hour and a zero appears in the database, the WECC path rating or the value set by the WECC OTC policy group is used.

Figure 1

% of Time exceeding x axis

% of OTC 75%

1%

Equivalent to the 99 percentile point

Max. flow Calculated

99% probability peak value

% of time exceeding 75% of OTC value

7

TABLE 1

OTC Values used in 2004-05 Path Flow Study

Path Number

Path Name OTC Value Comments (Source of OTC values)

1 Alberta to BC 0 to 735 MW EW (Ave

= 224 MW) 0 to 715 MW WE (Ave = 572 MW) (See OTC probability graph)

Hourly OTC from AESO, Hourly flows from BCH

3 Canada to NW 0 to 3150 MW NS

250 to 2950 MW SN Hourly OTC from BPA Hourly flows from BPA

8 Montana to NW 2200 MW E to W Constant

14 Idaho to NW 2400 MW E to W, 1200 MW W to E

Constant

15 Path 15 2050 to 5400 MW Posted and variable

17 Borah West 2307 MW Constant

18 Montana to Idaho 77-377 MW Posted and variable

19 Bridger West 371 to 2200 MW Posted and variable

20 Path C 300 to 950 MW Posted and variable

22 SW of 4 Corners 837 to 2325 MW Posted and variable

26 N to S California 3000 Max SN, 3135 MW Max NS

Posted and variable

30 TOT 1A 228 to 650 MW Posted and variable

31 TOT 2A 218 to 676 MW Posted and variable

34 TOT 2B 394 to 850 MW Posted and variable

35 TOT 2C 139 to 300 MW Posted and variable

36 TOT 3 450 to 1557 MW Posted and variable

8

45 CAISO - CFE 800 MW SN 408 MW NS

Constant Posted, but unchanging

47 S. New Mexico Max to 1038 MW Posted and variable

48 N. New Mexico Max to 1946 MW Posted and variable

49 E of River 2556 to 7550 MW Posted and variable

50 Cholla – Pinnacle Peak

1200 MW Constant

51 Southern Navajo 815 to 2264 MW Posted and variable

65 DC Intertie 0 to 3100 MW Hourly OTC from BPA & WECC

66 COI -3675 to +4800 MW Posted and variable

NOTES: “Posted and variable” – means the OTC value was posted hourly in the WECC EHV Data Pool database and is a varying quantity “Constant” – means the hourly OTC used in this analysis was a constant, unchanging value and was not posted in the WECC EHV Data Pool

Geographic location of the WECC Paths included in this analysis are shown in Figure 2. A description of the WECC Paths is contained in the WECC Path Rating Catalog.

9

10

SECTION III

PATH FLOW ANALYSIS AND RESULTS

Summary Graphs – 75% and 90% Indices

Figure 3 and 4 show in bar chart form the path loadings using the 75% and 90% indices. Results are presented in order of magnitude, from highest to least. Figure 3 (75% indices) covers the period 1998 thru 2005. Figure 4 (90% indices) covers the period 2004-05. Results for years prior to 2004-05 draw on analysis performed in those prior years. The 90% indices was not used consistently prior to 2004-05, hence the 90% analysis is only presented for the 2004-05 time period. For each path, the highest seasonal loading for the seasons analyzed is presented. Therefore, the three seasonal bars for each path may not be from the same year. For example, path 19 has the highest winter loading in 2000-01, the highest spring loading in 2001 and the highest summer loading in 2000. These are the values that are plotted for path 19. A line is drawn on the graph at the 50% and 25% levels. This was done to group the paths into loading ranges as an approximate measure of use, such as high, medium and low. This grouping was also used in presenting results on the Summary Path Loading Maps described below.

Summary Maps

Geographic path maps with path loading grouped into the following categories are shown in Figures 5, 6, and 7 for the 75% indices. Loadings are grouped on the maps into the following three categories:

• Paths with loadings greater than 75% of OTC occurring more than 50% of the time during a season

• Paths with loadings greater than 75% of OTC occurring between

25% and 50% of the time during a season

• Paths with loadings greater than 75% of OTC occurring less than 25% of the time during a season.

Separate maps are presented for each season showing the loading for the period from 1998 through 2005.

11

Individual Path Results – Seasonal 75% Indices 1998 thru 2005

The percentage of time each path exceeds 75% of the path OTC is plotted in the bar charts in the Appendix for each individual path from 1998 thru 2005, for the following seasons:

Winter – 1998-99, 2000-01, 2001-02, and 2004-05 Spring – 1999, 2001, 2002, 2004 and 2005 Summer – 1999, 2000, 2001, 2004 and 2005

These results draw on analysis performed in prior years. All seasons have not been analyzed in previous years, hence the missing seasons.

This represents the percentage of time during the season that the loading on the path exceeded 75% of the path transfer capability. For example, a path with a transfer capability of 1000 MW and a 30% calculated value for spring 2002 means that the path exceeded 750 MW (75% of OTC) for 439 hours (30% of a total of 1464 hours) during the months of April and May 2002. In some cases, a zero is reported for a path. This means that the path did not exceed 75% of the path OTC during the season. A path could have been operated at 74% of the path OTC for the entire season and the calculated value (for the % of time it exceeded 75% of OTC) would be zero.

Individual Path Results – Path Flow Frequency Distributions 2004-05

Frequency distribution plots for each path are shown in the Appendix for the Spring 2004 through Summer 2005 timeframe.

12

Figure 3

Path RankingMaximum Seasonal Values from 1999 through 2005

% of Time Path Actual Flow exceeds 75% of Path OTC

0

10

20

30

40

50

60

70

80

90

100

19 50 22 47 30 36 3 8 48 35 17 66 65 15 31 1 26 45 49 18 20 34 51 14

WECC Path Number

% T

ime

Path

Flo

w E

xcee

ds 7

5% o

f OTC

WSpSu

Represents the highest Seasonal Loading for each Path, from 1999 thru 2005

50% of Time

25% of Time

Bridger West

Cholla - Pinnacle Pk

SW of 4C

S. New Mex.

TOT 1A (CO to Utah)

TOT 3 (WY to CO)

13

Figure 4

Reliability Congestion Index - RCI (90%)Maximum Seasonal Values from Spring 2004 through Summer 2005

% of Time Path Actual Flow Exceeds 90% of Path OTC

0

5

10

15

20

25

19 3 1 65 66 30 22 36 45 15 35 18 31 48 26 20 14 17 47 8 50 49 34 51

Path Number

RC

I (90

%) -

% T

ime

> 90

% o

f Pat

h O

TC

90%

Ranking is based upon the Maximum 90% value for the five seasons between Spring 2004 and Summer 2005

Bridger West

NW to Canada (StoN)

Pacific DC Intertie

COI

TOT 1a (CO to Utah

SW of 4Corners

RCI (90%) is the percentage of time the physical flow exceeds 90% of the path OTC

Alberta to BC

14

Figure 5

15

Figure 6

16

Figure 7

17

SECTION IV

TREND ANALYSIS Path data has been analyzed for approximately a 7 year time period, from Winter 1998-99 through Summer 2005. An attempt was made to determine if there are any apparent path loading trends over this period of time. For example, is there a tendency for paths to be utilized or loaded more heavily now than earlier? Or are paths being utilized less heavily? Or is there no apparent trend? A number of factors could cause path loading trends to occur, such as increased or decreased line construction, changes in load growth, changes in generation type usage. Other factors may cause more random annual and seasonal variations, but less likely to affect longer-term trends, such as changes in hydro conditions, weather and temperature changes, etc. A very simple approach was taken to get a macroscopic view of the utilization of the entire interconnection to see if a trend existed. The method used was to simply sum the number of paths in the interconnection whose flows exceeded 75% of the path OTC for each season, from 1998 through 2005. Two summations were made to reflect the amount of time the paths exceeded the 75% value. Both summations are plotted in the associated figures below. The two were:

a) summation of the number of paths that exceeded 75% OTC greater than 50% of the time within each season

b) summation of the number of paths that exceeded 75% OTC greater than 25% of the time within each season

These sums were then plotted graphically in chronological order, from the Winter of 1998-99 through the Summer of 2005, to see if any visual trend was apparent. Results are plotted in Figure 8. The same data is replotted in Figures 9, 10 and 11 for the separate Winter, Spring and Summer seasons to remove the seasonal variations that are included in Figure 8. Recognizing the limited amount of data in the data sample (limited number of paths and years) and the simplified, quasi-quantified trending method used, it does appear that there may be a slight trend to lower utilization on a macro system basis with time for the years analyzed (between 1998 and 2005). This is more apparent in Figures 9, 10 and 11 where trends for each season are shown. This is particularly true for the summation for “greater than 50% of the time”. It should be emphasized that this analysis is on a macro total system basis and that individual areas of the system will exhibit different results. Other trending techniques should be investigated in the future to more accurately quantify these trends.

18

Figure 8

Path Utilization Trend1998 through 2005

Number of WECC Paths Exceeding 75% of their Path OTC, >25% and >50% of Time

0

1

2

3

4

5

6

7

8

9

10

W98-99 Sp 99 Su 99 Su 00 W00-01 Sp 01 Su 01 W01-02 Sp 02 Sp 04 Su 04 W04-05 Sp 05 Su 05Season

Num

ber o

f Pat

hs

25% Level50% Level

19

Figure 9

Path Utilization Trend - Winter Season1998 through 2005

0

1

2

3

4

5

6

W98-99 W00-01 W01-02 W04-05 W98-99 W00-01 W01-02 W04-05Season

Num

ber o

f Pat

hs >

75%

of p

ath

OTC

>25% of time>50% of time

20

Figure 10

Path Utilization Trend - Spring Season1999 thru 2005

0

1

2

3

4

5

6

7

8

Sp99 Sp01 Sp02 Sp04 Sp05 Sp99 Sp01 Sp02 Sp04 Sp05

Season

Numb

er o

f Pat

hs >7

5% o

f pat

h OT

C

>25% of time>50% of time

21

Figure 11

Path Utilization Trend - Summer Season1999 through 2005

0

1

2

3

4

5

6

7

8

9

10

Su99 Su00 Su01 Su04 Su05 Su99 Su00 Su01 Su04 Su05

Season

Num

ber o

f pat

hs >

75%

pat

h O

TC

>25% of time>50% of time

22

SECTION V

OBSERVATIONS AND CONCLUSIONS

The following observations may be drawn from the analysis:

1. The following paths had at least one season over the study period, in which the

seasonal loading exceeded 75% of OTC 50% of the time or greater: (These may be considered the more heavily utilized paths relative to their operating transfer capability. This by itself is not an indication that these are the most commercially congested paths. These are also not the most heavily loaded paths in terms of the magnitude of MW loading)

Path 19 – Bridger West Path 50 – Cholla – Pinnacle Peak Path 22 – Southwest of 4 Corners Path 47 – Southern New Mexico Path 30 – TOT 1A - (Colorado to Utah) Path 36 – TOT 3 - (Wyoming to Colorado) Path 27 – IPP DC Line (was not analyzed in this report, however it does meet

this criteria based upon previous studies.

2. Paths with the highest loadings relative to their transfer capabilities are primarily located in the Rocky Mountain and Desert Southwest regions (Wyoming, Colorado, Arizona and New Mexico).

3. Future analysis should include analysis of commercial data (ATC and schedules)

and correlation of observed hourly power flows with reported ATC and schedules.

4. Based upon the data analyzed in this report, it appears that total system loading

relative to total system capacity, may be decreasing slightly, however additional trend analysis should be performed to confirm this finding. It should be recognized that even if this is occurring, individual areas of the system might simultaneously be experiencing increasing transmission loadings and increasing transmission congestion.

5. As the loadings reflect power flow and not the schedules, or other contractual

uses of transmission that may not result in a flow (e.g. contracting transmission rights for forced outage reserves that are seldom used), it should not be concluded from this analysis that there was, or was not, significant congestion (defined as the inability to obtain transmission capacity when needed) on a path. Congestion analysis should include analysis of hourly ATC and hourly schedule data. In addition, no attempt was made to calculate the economic benefits of increasing the transfer capability. Additional studies would be needed to

23

determine if increasing the transfer capability of a path would bring economic benefits.