National Grid Performance Report

National Grid Performance Report i

Northeast Transmission Network Performance Report Fiscal Year 2005 National Grid Performance Report iii

Contents CONTENTS Transmission Reliability Page 4 Transmission and Customers Page 16 Developments Page 24 HVDC Page 32 Feedback and Additional Information Page 36

Welcome Transmission Reliability Performance Report This report has been prepared for National Grid s customers and other interested parties. The intention of this report is to provide its recipients information on National Grid s transmission reliability performance for Fiscal Year 2005 (April 1, 2004 through March 31, 2005). Performance trends over the previous years have also been provided. National Grid, through its operating subsidiary companies, owns and operates transmission assets in New England (in Massachusetts, Rhode Island and New Hampshire) and New York. In terms of this report, transmission is considered to consist of National Grid assets, operating at 69,000V and above in New England and 115,000V and above in New York, in these States only. This report provides information about the performance of National Grid s transmission system. National Grid USA is a registered public utility holding company headquartered in Westborough, Massachusetts. National Grid USA does not directly own or operate any facilities for the transmission or distribution of electricity; rather, such facilities are owned and/or operated directly by National Grid USA operating subsidiary companies. However, for ease of reference and for the convenience of the reader, the terms National Grid or Company shall be used in this document to refer to the transmission facilities and assets owned and/or operated by one or more of the subsidiary companies of National Grid USA. This document is made available with the expressed understanding that neither National Grid USA nor any of its affiliates, officers, directors, employees or agents assumes or makes any representation with respect to the contents, accuracy or completeness of this document. No part of this document may be copied, reprinted or reproduced without the permission of National Grid USA. Copyright National Grid 2005 2 National Grid Performance Report

Transmission Reliability Transmission Reliability National Grid Performance Report 3

Transmission Reliability 1 TRANSMISSION RELIABILITY This report has been produced principally for National Grid s transmission customers following positive feedback on documents from prior years. 1.1 Definition The North American Electric Reliability Council (NERC) defines reliability as the degree to which the performance of the elements of the system result in power being delivered to customers within accepted standards and in the amount desired. This definition contains the concepts of adequacy and security. Using NERC definitions again, adequacy is the ability of the system to supply the aggregate power and energy requirements of the consumers at all times and security is the ability of the system to withstand sudden disturbances. Transmission failures cause only a small percentage of power outages, yet, are significant in their impact on customers. This report provides information on the various metrics that we use to monitor and improve performance of our transmission system. These metrics measure reliability of the transmission system and consequently, the impact on customers. 1.2 Key Message Overall, system performance in is similar to our 5 year average. Although we have experienced a greater number of outages (compared to FY2004), generally the impact of those outages on customers was less severe, reflecting better performance. This performance in includes above average lightning activity which is a significant performance driver. 4 National Grid Performance Report

Transmission Reliability 1.3 Performance Summary The following table summarizes National Grid Transmission System reliability performance in compared to FY2004 and the 5 year Average. Table 1-1: Reliability Performance Reliability Check vs. FY2004 vs. Five Year total number of outages Up by 17% In Line with 5 year average number of momentary outages Up by 24% In Line With 5 year average number of sustained outages Up by 3% Worse than 5 year average number of loss of supply incidents Down by 4% Better than 5 year average number of loss of generation incidents Down by 19% Worse than 3 year average average duration of sustained outages Down by 6% Better than 5 year average index of transmission reliability Better by 3% In line with 5 year average The information presented in the above table is discussed in further detail in this report. 1.4 Selected Performance Statistics Table 1-2: Reliability Performance for National Grid in % of circuits that experienced no disturbances 48% % of circuits that only experienced a single disturbance 23% % of circuits that experienced 2 disturbances 12% % of circuits that experienced 3 or more disturbances 18% Overall ITR 99.963% Overall Availability 98.613% Average Transmission Circuit Interruption Duration Index (TSAIDI) Average Transmission Circuit Interruption Frequency Index (TSAIFI) 1.177 195 minutes Most customers will not have experienced loss of supply due to Transmission disturbances. 48% of our circuits had no disturbances at all and of those that did, most of those path interruptions did not result in a disturbance to customers. National Grid Performance Report 5

Transmission Reliability 1.5 Number of Disturbances The following figure compares transmission outages distributed through the year for both FY2004 and. The blue line shows the outages for FY2004 and the red line shows the outages for. These numbers represent the total number of disturbances, which were either momentary or sustained in nature. Figure 1-1: Cumulative Number of Outages by Month 700 600 500 400 300 200 100 0 Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar FY2004 5 YR AVG The number of disturbances in was higher than for FY2004. Although this represents a slight decrease in reliability performance over last year, performance was mostly consistent with the 5 year Average. This performance must be considered taking into account that the lightning season in FY2004 was significantly better than the 5 year average while that in was significantly worse. The highest rate of incidence of outages occurred during warm weather conditions with 58% of all outages occurring between May and August. Compared to FY2004, there was an increase of over 200% in lightning initiated outages in (an increase of 33% over the 5 Year Average). The following figure indicates the weekly profile of events in in both New England and New York. The green bars indicate the New England events while the blue bars indicate New York events. Callouts are used against weeks that had abnormally large number of outages and identify the major causes for the performance. In Lightning storms were responsible for all the major event weeks. 6 National Grid Performance Report

Transmission Reliability Figure 1-2: Weekly Number of Events in 60 Weekly Number of Incidents - 50 May 22-24, Lightning Number of Incidents 40 30 20 Jul 2, Lightning Aug 20-21, Lightning Aug 29, Lightning 10 0 Wk1 Wk3 Wk5 Wk7 Wk9 Wk11 Wk13 Wk15 Wk17 Wk19 Wk21 Wk23 Wk25 NE Wk27 Wk29 Wk31 Wk33 NY Wk35 Wk37 Wk39 Wk41 Wk43 Wk45 Wk47 Wk49 Wk51 Wk53 TSAIFI is the average number of interruptions per circuit. It is calculated by dividing the total number of disturbances in a given period by the number of circuits. TSAIFI for was 1.77. This indicates that each circuit had, on average, 1.177 disturbances last Year (equals 631 disturbances in total). Of these disturbances 433 were momentary outages and 198 were sustained outages. Figure 1-3: 5 Year Trend of TSAIFI TSAIFI 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0.0 FY2001 FY2002 FY2003 FY2004 With the exception of FY2004 which experienced a quiet lightning season, TSAIFI for is similar to performance in previous years and close to the 5 year average of 1.16. National Grid Performance Report 7

Transmission Reliability 1.6 Momentary Disturbances Momentary disturbances are measured through TMAIFI. TMAIFI is calculated by dividing the total number of momentary disturbances in any given period by the number of circuits. A lower value of TMAIFI represents better reliability performance. TMAIFI performance in was worse than in FY2004 reflecting an increase in the number of momentary outages. The highest incidence rate of momentary outages was in May and August contributing 40% of the total momentary outages for. Most of the outages in May and August were due to increased lightning experienced in those months. Initiatives to understand and improve performance with respect to lightning are ongoing. For more information on the Lightning initiatives, please refer to Section 3.1 Figure 1-4: 5 Year Trend of TMAIFI TMAIFI 1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 FY2001 FY2002 FY2003 FY2004 1.7 Sustained Disturbances Sustained disturbances are measured through TSAIFI-S. TSAIFI-S is computed by dividing the total number of sustained outages by the number of circuits. A lower value of TSAIFI-S indicates better reliability performance. The final value of TSAIFI-S for was higher than FY2004, showing a slight reduction in performance in. 8 National Grid Performance Report

Transmission Reliability Figure 1-5: 5 Year Trend of TSAIFI-S TSAIFI-S 0.45 0.40 0.35 0.30 0.25 0.20 0.15 0.10 0.05 0.00 FY2001 FY2002 FY2003 FY2004 1.8 Transmission Outage Duration The average duration of each sustained outage is measured through TSAIDI. TSAIDI is computed by dividing the total duration of outages by the number of circuits. For TSAIDI, measured in minutes, a lower value represents better reliability performance. TSAIDI was better in than for FY2004. The final value of TSAIDI for is 195 minutes, a slight decrease from last year. Figure 1-6: 5 Year Trend of TSAIDI TSAIDI 300 250 200 150 100 50 0 FY2001 FY2002 FY2003 FY2004 National Grid Performance Report 9

Transmission Reliability 1.9 Index of Transmission Reliability National Grid uses an index of transmission reliability (ITR) as a measure for assessing overall transmission reliability. It represents the percentage of time transmission circuits were available for the transmission of electricity from one terminal to another (or others). Thus ITR is a measure of unplanned unavailability of our transmission system. It is reduced by outage time due to unplanned outages (disturbances) on the system but it does not include outage time due to planned outages. As with other metrics we can categorize ITR by voltage classes and the following figure shows three ITR curves for FY2004 for Load (69 kv and 115 kv), Bulk (230 kv and 345 kv) and System (all circuits). Figure 1-7: Bulk, Load and System ITR for 100.00% 99.95% 99.90% 99.85% 99.80% ITR % 99.75% 99.70% 99.65% 99.60% 99.55% 99.50% Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Load ITR Bulk ITR System ITR Major reductions in both Load and Bulk Systems ITR have been mostly due to lightning. There were several events throughout the warm weather months when lightning caused significant damage to transmission structures, poles or conductors. Not all of these incidents impacted customers. In, 11% of all outages on the transmission system directly impacted customers. 10 National Grid Performance Report

Transmission Reliability 1.10 System Availability Availability is calculated based on the total length of planned, unplanned, and disturbance durations and the total hours the system or circuit should have been available in a given period. The availability of the system measures the increased reliability of the system and is important in case of an 'unplanned' event. High availability provides fewer opportunities and less compromising situations in the event of failure when back up (redundant circuits) are in service (available). The potential risk is higher for line and customer related outages when all circuits are not available. In order to maintain a high level of Availability, circuits are returned to service as quickly as possible in the event of unplanned disturbances. In the case of planned outages which require work over several days, circuits are returned to service at the end of each work day. The value of System Availability for 98.63 for New England and 98.60 for New York. The following figure indicates the availability of the Load, Bulk and the Transmission System in. Figure 1-8: Bulk, Load and System Availability for 100.00% 99.00% 98.00% 97.00% 96.00% 95.00% 94.00% Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Load Bulk System System FYTD National Grid Performance Report 11

Transmission Reliability 1.11 Outage Causes For every outage seen on the transmission system outage causes are identified and recorded. These causes are analyzed to identify trends and similarities. These studies provide insight into what could be done to possibly prevent future occurrences. This section contains information about the causes of transmission outages. Causes of both momentary and sustained outages have been indicated. The numbers of outages attributed to the different outage causes for All Outages, Momentary Outages and Sustained Outages have been indicated in the following figures. Figure 1-9: Causes of All Outages in Substation Equipment + System Protection Weather External Line Equipment Lightning All Others A significant number of outages (64%) were a result of Lightning and causes falling under the category of All Other. The category of All Other includes vandalism, problems with trees and other vegetation, and some for which the cause is unknown. In addition to line and substation issues, problems outside National Grid s control, classified as cause External contributed to 26% of the outages. The category Unknown includes outages whose cause could not be identified. These outages are mostly momentary in nature. Outages are classified as unknown only after foot and helicopter patrols reveal no valid explanation for the operation. The category External is assigned to outages that arise due to factors external to and outside the control of the system. These include foreign utilities, customer equipment, vehicles hitting equipment and generation. The number of outages initiated by External causes saw an increase of over 200% compared to FY2004. 12 National Grid Performance Report

Transmission Reliability 1.11.1 Causes of Momentary Outages Figure 1-10: Causes of Momentary Outages in Causes of Momentary Outages in Substation Equipment + System Protection Weather External Line Equipment Lightning All Others The preceding figure shows the number of momentary outages attributed to the various causes in. 84% of momentary outages were caused by Lightning, Weather or Unknown causes. Unknown causes form a part of the All Other category. Compared to FY2004, saw an increase of 261% in the number of momentaries due to Lightning, a direct reflection of the increased lightning seen in. National Grid Performance Report 13

Transmission Reliability 1.11.2 Causes of Sustained Outages It is interesting to look at the breakdown of sustained outages by cause and to compare these with the breakdown of momentary outages by cause. Figure 1-11: Causes of Sustained Outages in Causes of Sustained Outages in Substation Equipment + System Protection Weather External Line Equipment Lightning All Others The distribution of sustained outages causes shows that in addition to Lightning and Other causes, line and substation issues are also major drivers. 14% of the sustained outages were due to Substation and Line Equipment issues. Compared to FY2004, the number of sustained outages in due to External factors has increased by 233%. 14 National Grid Performance Report

Transmission and Customers Transmission and Customers National Grid Performance Report 15

Transmission and Customers 2 TRANSMISSION AND CUSTOMERS This section focuses on the impact of transmission performance on its customers. There are several types of transmission customers. Some may be affected by momentary outages for a few seconds, while some may suffer a prolonged disturbance due to a sustained outage. Loss of generation events impact system reliability directly through transmission impact and also contingency effects; both serve to disrupt the wholesale market. When transmission or distribution customers lose supply due to a sustained transmission outage, it is considered a loss of supply event. Loss of Supply and Loss of generation events impact customers and are therefore given particularly high priority. These events are analyzed closely and all possible efforts are taken to minimize their occurrence. 2.1 Loss of Generation For, there were 20 Loss of Generation incidents, 10 in New England and 10 in New York. Loss of Generation applies only if a generator is online at the time of the outage. All loss of generation events are captured regardless of the size of generation lost. In, the Losses in Generation ranged from 3 MW upwards. The following figure indicates the number of Loss of Generation events in. This trend is compared against that of FY2004. The improvement in over FY2004 is 23%. Figure 2-1: Cumulative LOG Incidents - vs. FY2004 30 25 Number of LOG Incidents 20 15 10 5 0 Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar FY2004 16 National Grid Performance Report

Transmission and Customers The following figure provides information on the causes of the various outages that resulted in loss of generation. The biggest contributor is the category of All Other. This includes problems due to vegetation, animals, human error and other unknown reasons. Figure 2-2: Causes of LOG Incidents in Causes of LOG Incidents Substation Equipment + System Protection Weather External Line Equipment Lightning All Others National Grid Performance Report 17

Transmission and Customers Some of the Loss of Generation events in New England and New York have been indicated below. In each case, the generator connected to the line was generating at the time of the trip. Date Circuit Details Impact 05/07/04 333 (Burriville, RI) Relay problem at Sherman Road Station caused the line and the connected generator to trip Loss of 200MW. 06/08/04 Feura Bush- North Catskill #2 (South of Albany, NY) The line tripped and reclosed when there were severe thunderstorms in the area. Loss of approximately 72 MW. 08/11/04 08/20/04 and 08/21/04 09/02/04 01/23/05 Spier- Queensbury #5 and #7 (Glens Falls, NY) 338 (Between Tewksbury and Woburn, MA) Spier Falls Bus (Saratoga County, North of Albany, NY) U6 (Between Somerset and Bridgewater, MA) The lines tripped and reclosed when there were severe thunderstorms in the area National Grid experienced numerous line operations due to severe thunderstorms. During the storm, the 338 line was interrupted. Breaker overhaul work at the Spier Falls Hydro plant, generator unit #9, caused the 115 kv 99G bus at Spier Falls station to trip/reclose auto The cause of the outage was a static wire on the U6 / V5 structure south of Taunton River in Dighton. An outage was required for V5 line in order to make repairs to U6 / V5 static wire. Loss of 130 MW. The 338 Line trip caused the loss of approximately 1,000 MW of generation in the Boston Area. This resulted in ISO-NE implementing Emergency Operating Procedures, including a 5% Voltage reduction in the Boston Area Loss of 133MW. Loss of 170MW. 18 National Grid Performance Report

Transmission and Customers 2.2 Loss of Supply For, there were 21 Loss of supply incidents in New England and 32 Loss of Supply events in New York. On average, in, a customer in New England would have lost supply for 9.2 minutes and a customer in New York would have lost supply for 3.5 minutes as a result of Transmission outages. The following figure indicates the trend of LOS events in New England across and compares it against that of FY2004. Figure 2-3: Cumulative LOS Incidents in New England 25 20 Number of LOS Incidents 15 10 5 0 Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar FY04 FY05 National Grid Performance Report 19

Transmission and Customers The following figure examines the causes of the various LOS Incidents in New England. The number of events due to weather, External and All Other causes all saw increases when compared to FY2004. Figure 2-4: Causes of LOS Incidents in New England in NE - Causes of LOS Events Substation Equipment + System Protection Weather External Line Equipment Lightning All Others Some of the significant New England LOS events have been indicated here: Table 2-1: Significant New England Loss of Supply Events for Date Circuit Details Impact 11/28/04 12/27/04 07/22/04 Q169 and A179 (Lynn, MA) M13 and L14 (Between Fall River, MA and Portsmouth, RI) E20, L1 and M1 (Between East Bridgewater, MA and Middleborough, MA) A downed Q169 Structure struck the guy wire on the A179 Line Icing on the M13 and L14 lines combined with strong winds created a galloping condition, causing the phases to come into contact with each other. Approximately 12 to 15 spans were observed galloping to different degrees. C phase of the L1 line down between structures #1 and #2. This action also de-energized the feed into Middleboro Municipal off the M1 line Approximately 20,000 customers lost power for up to a maximum of 678 minutes. A total of 65,412 customers were affected for up to 55 minutes. Approximately 22,260 customers lost power for a maximum of 71 minutes. 20 National Grid Performance Report

Transmission and Customers The following figure illustrates the distribution of the New York Loss of Supply events throughout compared against FY2004. The number of events in decreased indicating performance improvement. Figure 2-5: Cumulative LOS Incidents in New York 40 35 Number of LOS Incidents 30 25 20 15 10 5 0 Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar FY2004 The following figure examines the various causes of New York Loss of Supply Incidents in. While there was a decrease in the number of events due to equipment issues, those initiated by lightning, External and All Other categories increased. Figure 2-6: Causes of LOS Incidents in New York in NY - Causes of LOS Events Substation Equipment + System Protection Weather External Line Equipment Lightning All Others National Grid Performance Report 21

Transmission and Customers Some of the significant New York LOS events have been indicated here: Table 2-2: Significant New York Loss of Supply Events for Date Circuit Details Customer Impact 05/23/04 01/24/05 08/17/04 Levitt-Rome #8 (West of Utica, NY) Thousand Islands-Coffeen #4 (Watertown, NY) Republic- Ticonderoga #2 (Western shore of Lake Champlain in New York) The line tripped due to a lightning strike and locked out due to a stuck relay contact Coffeen station 115 kv Capacitor Bank was removed from service and subsequently the capacitor bank circuit interrupter exploded. Several breakers opened and reclosed automatically but one of them failed to reclose due to a burnt closing coil, causing a sustained outage to the Thousand Islands-Coffeen #4 line. The trip was caused when a guy wire broke and got up into the conductor Approximately 6000 customers were interrupted for11 hours and 15 minutes. Approximately 15,000 customers were interrupted about 1 hour. Approximately 4400 customers lost power for 131 minutes. The event also caused loss of generation of 44MW. 22 National Grid Performance Report

Developments Developments National Grid Performance Report 23

Developments 3 DEVELOPMENTS 3.1 Current and Future Assignments Several projects were implemented in order to improve transmission reliability as well as service to customers. Some of them have been outlined here: 3.1.1 Rehab Projects A wide range of Line Rehab projects were implemented in. Figure 3-1 shows New England projects occurring in. The line numbers/ circuits are a relative indication of where the circuit is physically located and not necessarily the location of the construction. These construction projects include many aspects of line rehabilitation including reconfiguration, reconductoring switch installations and replacements. Figure 3-1: Projects in New England in 24 National Grid Performance Report

Developments Figure 3-2 shows New York projects occurring in. These construction projects focus on a variety of rehabilitation issues. Multiple line numbers are listed in a single call-out instead of using multiple call-outs to avoid cluttering the area map. The circuit numbers are cross-referenced below with the typical New York standard circuit naming convention. Figure 3-2: Projects in New York in Circuit # Circuit Name Circuit Name T1280 Gardenville - Homer Hill #152 T2660 Teall - Carr Street #6 T1310 Gardenville - Homer Hill #151 T2670 Teall - Oneida #2 T1440 Huntley - Lockport #36 T3060 Black River - Taylorville #2 T1450 Huntley - Lockport #37 T3080 Browns Falls - Taylorville #3 T1500 Lockport - Batavia #108 T3130 Coffeen - West Adams #2 T1530 Lockport - Mortimer #111 T3320 Taylorville - Boonville #5 T1540 Lockport - Mortimer #113 T3330 Taylorville - Boonville #6 T1550 Lockport - Mortimer #114 T3340 Taylorville - Moshier #7 T1580 Mortimer - Golah #110 T4140 Oneida - Oneida Energy (Sterling) #4 T1720 Niagara-Packard #62 T4210 Porter-Rotterdam #31 T1810 Packard - Walck Road #129 T5010 Albany - Greenbush #1 T1820 Packard - Huntley #130 T5020 Albany - Greenbush #2 T1890 Southeast Batavia - Golah #119 T5060 Battenkill - North Troy #10 T2000 Ash - Teall #7 T5450 New Scotland - Alps #2 T2010 Ash - Teall #8 T5530 North Troy - Hoosick #5 T2140 Curtis Street - Teall #13 T5630 Rotterdam-Bear Swamp E205 T2200 Elbridge - Woodard #4 T5680 Rotterdam - New Scotland #13 T2480 Oswego - Volney #12 T6020 Walck Road - Huntley #133 T2580 South Oswego - Curtis St. #10 T6270 North Carthage - Taylorville #8 National Grid Performance Report 25

Developments 3.1.2 Lightning Analysis As indicated in the preceding figures and tables, lightning has a very significant impact on the system performance. Recognizing this importance, a lightning improvement initiative was established in late. There are two interrelated aspects to this work: Lightning Analysis Lightning Mitigating Solutions The overall intention is to learn more on how lightning impacts operational performance. A systematic analysis of historical faults (over the past 5 years) is being performed to derive a list of circuits that are especially prone to and exhibit poor performance with respect to lightning. The intent is then to develop strategies and projects to address this poor or undesirable performance. In addition, strike location information is being identified for lightning-caused faults, to facilitate targeting and identification of potential damage. This would facilitate shorter repair times as well as mitigate the potential for future outages caused by lightning. Historically, northeastern United States is an area with relatively lower long-term exposure to lightning when compared to many other areas of the country, but with this lower exposure comes a much greater variability in the amount of lightning storms from season to season. In the ten years studied for the National Grid USA service area average exposure densities (lightning density is the number of ground strokes/square kilometer/year) variances of ± 50 to 100% from the long-term average for a year are common, and some year-to-year changes are as high as 500%. 26 National Grid Performance Report

Developments The lightning density maps below illustrate the drastic nature of this year-to-year variability. Compare the density maps for CY2003 and CY2004. Hotter colors indicate higher densities of lightning ground strikes. National Grid Performance Report 27

Developments This chart shows the last five fiscal years cumulative lightning exposure, broken into regions. Note that the average cumulative exposure for this period was approximately 320,000 strokes. Cumulative Lightning Exposure - fiscal Years 2001-2005 800 700 600 Strokes (Thousands) 500 400 300 200 100 0 2001 2002 2003 2004 2005 NE NY The large variability in exposure to lighting from season to season presents a significant challenge to the lightning analysis program in verifying whether or not the mitigation strategies that are eventually implemented have been effective. Since the individual circuit or section of circuit that was improved may not be challenged in that area again for some years, it may take as long to check that the improvement was correctly targeted. This places a premium on the careful selection of circuits and circuit weakness areas in the various stages of the program. The details on the two major aspects of the Lightning Improvement program are outlined below: 3.1.2.1 Lightning Analysis Strategies The initial stages of the program would involve screening and selection of circuits. The Challenge/Fault ratio would be used to develop relative rankings of circuits according to their lightning performance, thereby identifying the poorest performers in each region. Each succeeding stage following the initial screens will take a great deal more time on each circuit, so by starting with a smaller population of underperformers, the overall analysis time is reduced. The next logical step in the program would be to identify circuits that should exhibit reliability performance improvement if changes are made. At this point, a broader analysis of engineering efficacy, business need, and projected circuit performance would be performed on the list of worst lightning performers to come up with candidates for mitigation projects. 28 National Grid Performance Report

Developments Some of the additional criteria that will be employed are: The projected fault rate calculated using the expected long-term exposure from the regional lightning density and the circuit s Challenge/Fault Ratio Soil Resistivity/Grounding Structure Types Elevation/Geography 3.1.2.2 Lightning Mitigating Strategies This stage will involve the engineering of mitigation strategies to be used as addons to existing projects or for future projects. Some possible solutions that will be considered: Improved Grounding Counterpoise Insulator Upgrades/Replacement Shield Wire Lightning Arrestors All circuits where mitigation projects are implemented will continue to be monitored to track their lightning performance improvement. 3.1.3 Vegetation Management Program National Grid s Transmission Forestry Department implemented a plan to carry out a risk assessment study and develop a mitigation strategy on its NE (69 kv, 230 kv and 345 kv) and NY (115 kv, 230 kv and 345 kv) transmission system. The risk assessment was focused on quantifying the size and characteristics of the utility forest outside the electric utility right-of-way, this being the primary source of transmission system interruption/loss of supply across the system. The goal of the Transmission Forestry Department is to minimize this risk, balanced against financial resources, to improve overall system reliability as measured by the number of tree caused incidents and loss of supply to customers. Within this goal, there is a particular focus on major transmission lines (230kV and 345 kv) where National Grid seeks a management regime that results in no tree-caused service interruptions. The project was designed to provide the Transmission Forestry Department the data required to quantify the current level of tree risk. Risk of outages from the utility forest adjacent to cleared rights-of-way is primarily related to three factors: cleared width (distance from conductor to tree line), conductor height above ground and tree height. The Optimal Width Calculator developed by Ecological Solutions, Inc., a consulting firm contracted by National Grid, is used to calculate a Risk Factor to quantify risk of a tree falling into National Grid Performance Report 29

Developments conductors based on the three factors. The Risk Factor is a relative risk rating the numeric value has no direct meaning except in relation to other data. 3.1.4 Asset Management Process The objective of the Transmission Asset Management Business Model was to create a process based on an asset steward and a service provider model through which the Transmission and Distribution Business segments will be responsible for the stewardship of their respective allocated assets, while receiving services from organizations in other parts of the business. The Transmission Network Asset Management Process is moving from a basic approach to a progressive approach in implementation of Network Review, Strategic Options, Work Program Development, and Work Program Delivery. The objective is to combine understanding of the current condition and performance of all the network assets making up the National Grid USA Northeast Transmission System. It is intended to collate data, information and knowledge from a number of sources across the transmission business in order to provide a balanced view of the issues affecting equipment condition and performance, and recommend remedial actions. The current status of these assets would be identified, highlighting any risks or opportunities that the condition or performance of the asset represents to meeting the business objectives, now and into the future. 3.2 Business Initiatives The National Grid Five Year Statement, 2005-2009 provides insights into generation development and transmission system loadings and limitations across the New York and New England region. It also explores options for the reinforcement of the transmission system, and indicates where new or expanding generation or load can best be accommodated. The Five Year Statement is available on request. 30 National Grid Performance Report

HVDC HVDC National Grid Performance Report 31

HVDC 4 HVDC This section contains information relating to the performance of the 450 kv DC facility terminating at Sandy Pond. National Grid USA operates the New England portion of two interconnections know as Phase 1 and Phase 2, between New England and Canada. Phase 1 consists of two converter stations (Des Cantons in Quebec, and Monroe in NH), each rated 690 MW. Phase I was placed in commercial operation service in October 1, 1986. Phase 2 included three additional converter terminals (Radisson, Nicolet and Sandy Pond). Sandy Pond is a + 450 kv DC 2000 MW bipolar converter terminal located in Ayer, MA. Phase 2 (Radisson and Sandy Pond only) began commercial operation on November 1, 1990. In 1992 Nicolet, the third terminal on the multi-terminal HVDC system was placed in service. This terminal is rated 2138 MW and located in the Montreal area. Phase I and Phase II share the same dc transmission lines. As a result Sandy Pond and Monroe may not operate simultaneously. Sandy Pond is the preferred station of operation due to higher capacity rating (2000MW vs. 690MW) and it is located closer to larger load centers. Figure 8-1: HVDC Connection between James Bay and Sandy Pond 32 National Grid Performance Report

HVDC Most HVDC transmissions are point to point transmissions using overhead lines. Most HVDC overhead line transmissions are bipolar, i.e. they use two conductors of opposite polarity (one positive and one negative). A bipolar transmission is in fact a double circuit transmission, since one pole can continue to transmit power when the other pole is out of service. A multi-terminal HVDC transmission is more complex than an ordinary point-to-point transmission. In particular, the control system is more elaborate and the communication requirements between the stations become larger. There is only one large-scale multi-terminal HVDC system in operation in the world today. It is the 2000 MW Hydro Québec Sandy Pond interconnection. The operating experience of this transmission has proved that from a technical point of view there are no problems connecting several converter stations to the same HVDC line. 4.1 Utilization Phase 1 is rarely used to allow the use of Phase 2. Phase 2 had a Firm Energy Contract that expired end of August 2001. This can be seen in the reduction in Phase 2 utilization during FY2002. There were other commercial issues ongoing at this same time that may have contributed to lower utilization. Since that time utilization has been back up in the mild and warm months and generally decreased in the winter time period. Hydro Quebec s peak is in the winter and energy is typically diverted to Nicolet when it is very cold rather than sending it south to New England. The following figure shows the utilization of Phase 2 for each month during. Figure 8-2: Sandy Pond Monthly Utilization in 45% Sandy Pond (Phase 2) Utilization 40% 35% 30% 25% 20% 15% 10% 5% 0% Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar National Grid Performance Report 33

HVDC 4.2 Phase 2 Operational Issues There is a large fire protection project underway at Sandy Pond. The VESDA (Very Early Smoke Detection and Alarming) System used as part of the Valve Hall fire protection system is required. The present VESDA System s smoke detectors no longer have direct replacements available. Initiatives on his project and other asset strategies for Sandy Pond are managed by Transmission Asset Strategy Department. While HVDC technical personnel continue to support the facility, availability of replacement parts and expertise are important issues. 34 National Grid Performance Report

Feedback and Additional Information Feedback and Additional Information National Grid Performance Report 35

Feedback and Additional Information 5 FEEDBACK AND ADDITIONAL INFORMATION 5.1 System Information Table 5-1: Asset Information for National Grid # of circuits 536 miles of transmission included for statistical reporting 8,767 Number of circuits / miles at 69 kv 41 / 537 Number of circuits / miles at 115 kv 415 / 6,070 Number of circuits / miles at 230 kv 31 / 945 Number of circuits / miles at 345 kv 49 / 1,215 5.2 Definitions / Acronyms A number of commonly used terms are used in this report. These and other definitions that the reader may be familiar with are described here. Availability Bulk Transmission Circuit Circuit Mile Customer Delivery Point Disturbance Percentage of time the entire transmission system (or a particular class or circuit) is available for the transmission of electricity between two or more terminals. It is calculated based on the total length of planned, unplanned, and disturbance durations and the total hours the system or circuit should have been available in a given period. The highest attainable value would be 100%. A voltage class under study and composed of 230 kv or higher voltage transmission circuits (excludes HVDC). An electrical asset that is in place for the purpose of transmitting or delivering electrical energy between two or more terminals. A circuit is composed of a breaker-to-breaker or breaker to load segment of the transmission grid. For transmission, the circuit must be energized at a phase-to-phase voltage of 69 kv or higher. The total length of a designated circuit in miles. This would be the sum of all the span lengths (pole, tower, etc.), in a given circuit. The ultimate consumer of electricity supplied by the transmission system. A customer can be residential, commercial, industrial, or municipal. Currently, municipalities served by the transmission system are considered one customer each. A delivery point is the location where the transmission system delivers electricity to the distribution system. The National Grid defined delivery point is the low-side bushings of the transformer or metering point. An event that occurs to a circuit(s), which results in the automatic operation (open, open-close, open-close-open) of one or more terminals (circuit breakers, sectionalizing devices) of that circuit. The voltage on the circuit DOES NOT need to drop to zero to be considered a disturbance. Customers may or may not be interrupted by the disturbance. These events are forced, unplanned incidents, usually because of lightning, weather, substation trouble, animal contact or other unanticipated influences. 36 National Grid Performance Report

Feedback and Additional Information Disturbance Duration Fault HVDC Interconnection Interruption ITR KPI LCM Line Load Transmission LOG LOS Momentary Disturbance Outage The total time a circuit is not available for power flow to an interconnection, delivery point or between two or more terminals. The duration is measured in minutes and is the time difference from when a circuit is interrupted from normal power flow and when it is restored to normal power flow. A phase-to-phase or phase-to-ground condition on a circuit, which may result in the operation of the protective devices of the circuit. A fault may be transitory and clear before the first re-close of a circuit, or result in a lock out condition. High Voltage Direct Current. A method of transmitting electrical energy using the Direct Current (DC) method as opposed to the Alternating Current (AC) method. A terminal is in place at each end of the line for converting from AC to DC and vice versa. The HVDC line from James Bay to Sandy Pond allows for import or export of approximately 2 GW from either terminal. An interconnection is a point where transmission service from the transmission provider is transferred to another entity (ex. National Grid to NStar). These are normally closed points of service to generators, nonaffiliated utilities, wholesale customers, municipalities, etc. One or more openings of a protective device, resulting in zero voltage or discontinuity, occurring on a single circuit or on adjacent connected facilities. (Not consistent with IEEE definition by which an interruption is an outage that causes a customer to lose supply.) The Index of Transmission Reliability is the percentage of time the entire transmission system (or a particular circuit) is available for the transmission of electricity from one terminal to another or others. Key Performance Indicator. A metric that represents an important facet of transmission system performance. Lost Customer Minutes is a product of the estimated number of customers without supply multiplied by the duration of the interruption. Where multiple feeders are involved the LCM is the sum of the products of these calculations for each feeder. See Circuit. The two terms are used interchangeably. A voltage class under study and composed of 69 kv and 115 kv transmission circuits. Loss of Generation. A disturbance on the transmission system, which results in the inability of a generation facility to produce and transfer energy. The cause of this disturbance can be initiated by either a distribution or transmission event, but not the generator itself. Loss of Supply. A sustained disturbance on the transmission system, which results in the loss of electrical service to one or more customers. The cause of this disturbance can be initiated either by a distribution or transmission event, with the cause being noted and tracked. A disturbance whose total duration is less than or equal to one minute for New England and New York. Multiple operations of a protective device followed by a successful re-close within the defined momentary duration above would be considered one momentary disturbance. Interruption of a transmission circuit, either by a disturbance or planned or unplanned maintenance operation, which causes the circuit to become unavailable for normal power flow. National Grid Performance Report 37

Feedback and Additional Information Sustained Disturbance TCI Terminal TMAIFI TSAIDI TSAIFI TSAIFI-S Voltage Class A disturbance whose total duration is greater than one minute for New England and New York. Multiple operation of a protective device followed by a lock out would be considered one sustained disturbance. Transmission Customer Impact is a measure of loss of supply and/or loss of service events to customers (including generators). A score of 0 is the best level of performance. It is derived from a simple count of events that cause a loss of supply / loss of access to the system. A disturbance that results in both LOS and LOG counts as one TCI. A point where transmission line conductors terminate and they are connected to one or more sectionalizing devices. The purpose of the terminal is to isolate the given circuit. Typically, a terminal would be in a substation or switch station, where sectionalizing devices are also connected to a station bus or transformer. Transmission Momentary Average Interruption Frequency Index. A value, which is calculated by dividing the total momentary disturbances in a given period by the number of circuits in the system. Multiple operations of a protective device followed by a successful re-close within a total duration of 1 minute or less would be considered one momentary disturbance. Transmission System Average Interruption Duration Index. Total duration of circuit interruptions in a given period divided by the number of circuits in the system. Note: Does not conform explicitly to the IEEE SAIDI definition. Transmission System Average Interruption Frequency Index. The average interruptions per circuit. It is calculated by dividing the total number of interruptions by the number of circuits in the system. Note: Does not conform explicitly to the IEEE SAIFI definition. Transmission System Average Interruption Frequency Index - Sustained. It is computed by dividing the total number of sustained (>60 seconds unless otherwise stated) interruptions by the number of circuits in the system. A voltage level used for analysis purposes, by grouping together circuits with a common transmission voltage for transmission performance reporting. The AC transmission voltages under analysis are 69 kv, 115 kv, 230 kv, 345 kv. Also DC transmission voltage of 450kV (not used in this report). 5.3 Feedback Information As part of our effort to improve both reliability performance and customer satisfaction we welcome your feedback. We hope that this report has provided both useful and interesting information about our performance and about our continuing efforts to make improvements. Please contact your Transmission Account Manager to share your views and comments. If you need additional information on this Report, please contact: Herb Schrayshuen Transmission Commercial Services National Grid USA 300 Erie Boulevard West Syracuse, NY 13202 Email herbert.schrayshuen@us.ngrid.com 38 National Grid Performance Report

Feedback and Additional Information