July 11, 2013 Reliability and Resilience – Role of Smart Grid Technologies and Practices Joe Paladino A National Town Meeting on Demand Response and Smart Grid, Washington, DC, July 9 - 11 1

Recovery Act Grid Modernization Programs Smart Grid Investment Grants (SGIG)* $3.4 billion Smart Grid Regional Demonstrations (SGDP)* $620 million Workforce Training $100 million Interconnection-wide Transmission Planning and Resource Analysis $80 million Interoperability Standards (with NIST) $12 million Other Technical Assistance to States on Electricity Policy ($44 million) Local Energy Assurance Planning ($10 million) One-time Appropriation $4.5B of Recovery Act Funds Investment Grants Smart Grid Demos Workforce Training Resource Assessment & Transmission Planning Smart Grid Interoperability Standards Other Smart Grid Demos Workforce Training Resource Assessment & Transmission Planning Smart Grid Interoperability Standards Other One of the requirements of the Recovery Act is that funds for SGIG, SGDP, and Workforce Training had to be matched with cost sharing at least dollar for dollar. For example, the $3.4 billion for SGIG is being matched with $4.4 billion in private funds for a total budget of $7.8 billion. In total, these funds are supporting more than 180 smart grid projects. *Originally authorized by the Energy Infrastructure Security Act 2007, EISA 1306 and EISA 1304 2

Applications and Benefits Matrix Smart Grid Technology Applications Consumer-Based Demand Management Programs (AMI-Enabled) Advanced Metering Infrastructure (AMI) Applied to Operations Fault Location, Isolation and Service Restoration Equipment Health Monitoring Improved Volt/VAR Management Synchrophasor Technology Applications Time-based pricing Customer devices (information and control systems) Direct load control (does not require AMI) * *Shifting peak demand may or may not lower emissions Meter services Outage management Volt-VAR management Tamper detection Back-Office systems support (e.g., billing and customer service) Automated feeder switching Fault location AMI and outage management Condition-based maintenance Stress reduction on equipment Peak demand reduction Conservation Voltage Reduction Reactive power compensation Real-time and off-line applications Capital expenditure reduction – enhanced utilization of G,T & D assets ✔ Energy use reduction Reliability improvements O&M cost savings Reduced electricity costs to consumers ✔* Lower pollutant emissions Enhanced system flexibility – to meet resiliency needs and accommodate all generation and demand resources This table provides a more granular look at the benefits from various smart grid technology applications. The main point is to underscore the need for careful accounting of all benefits in evaluating projects and investments. Not shown but also essential in evaluating smart grid investments are the serious implementation issues that are encountered in moving from pilot projects involving relatively small numbers of customers and feeders to system-wide or near system-wide deployments. These issues include systems integration and interoperability between new and legacy systems, cybersecurity (including devices, systems, and business practices), and customer acceptance (including data privacy.) Also important are appropriate mechanisms for cost recovery so that financial risks can be properly assessed. 3

Recent DOE SGIG Reports Analysis of Customer Enrollment Patterns in Time-Based Rate Programs June 2013 Economic Impact of Smart Grid Investments April 2013 2012 Progress Report July 2012 Impact Reports December 2012 O&M Savings from AMI Demand Reductions from AMI Volt/VAR Optimization Reliability Improvements from DA As of May 2013 Download From We invite everyone to visit smartgrid.gov and download these documents. The website also contains quarterly updates on equipment installations and expenditures and case studies highlighting early findings and lessons learned from about 20 selected projects. In the Consumer Behavior Study section there are specific lessons learned pages from implementing time-based rate programs in conjunction with AMI and customer systems. The content is being regularly updated and there are new reports and case studies being produced. For example: LBNL is publishing two reports on the experimental designs of the CBVS projects and initial results on customer recruitment efforts from the studies. ORNL is publishing a report comparable to the four impact studies shown on the slide with initial results for the ten SGIG synchrophasor projects. LBNL, ORNL, and Navigant Consulting are working on additional reports and analysis as more information on the results of the projects become available. We are building an email list and working with NARUC and other stakeholders to find ways of letting everyone know when new information becomes available.

Recent Weather Patterns Between 2003 and 2012, over 670 widespread weather related power outages occurred Extreme weather related events have become more frequent and/or more intense over the time period. The 2012 events included 11 weather and climate disaster events – 7 severe weather/tornado events, 2 tropical storm/hurricane events, and the year-long drought and associated wildfires – second only to 2011, when 14 events were recorded. 2012 was the 2nd costliest year on record; (11 disaster events cumulatively caused over $110 billion in damages (Superstorm Sandy cost approx $65 billion). Sandy Storm Surge was 14.06 feet (8 feet higher than highest tide).; water rose at a rate of 1-3’ per 6 minute at the most intense rate 5

Reliability Improvements from Automated Feeder Switching Selected examples from SGIG projects reporting initial results 4 Projects involving 1,250 feeders April 1, 2011 through March 31, 2012 Index* Description Weighted Average (Range) SAIFI System Average Interruption Frequency Index (outages) -22 % (-11% to -49%) MAIFI Momentary Average Interruption Frequency Index (interruptions) (-13% to -35%) SAIDI System Average Interruption Duration Index (minutes) -18 % (+4% to -56%) CAIDI Customer Average Interruption Duration Index (minutes) +8 % (+29% to -15%) 6

Value of Service One utility has installed 230 automated feeder switches on 75 circuits in an urban area. From Apr 1 – Sep 30 2011: SAIDI improved 24%; average outage duration decreased from 72.3 minutes to 54.6 minutes (or by 17.7 minutes). Estimated Avg. Customer Interruption Costs US 2008$ by Customer Type and Duration Customer Type Interruption Cost Summer Weekday Interruption Duration Momentary 30 mins 1 hr 4 hr 8 hr Large C&I Cost Per Average kWh $173 $38 $25 $18 $14 Small C&I $2,401 $556 $373 $307 $2,173 Residential $21.6 $4.4 $2.6 $1.3 $0.9 Sullivan J, Michael, 2009 Estimated Value of Service Reliability for Electric Utility Customers in the US, xxi VOS Improvement Δ = Δ SAIDI x Customers Served x Avg Load x VOS Coefficient VOS Estimate for SAIDI Improvement on 75 feeders from Apr 1 to Sep 30 2011 Customer Class Δ SAIDI Customers Served within a Class Average Load (kW) Not Served VOS Coefficient ($/kWh) Δ VOS Residential 17.7 mins (0.295 hrs) 107,390 2 $ 2.60 $ 164,736 Commercial 8,261 20 $ 373.00 $ 18,179,477 Industrial 2,360 200 $ 25.00 $ 3,481,325 Total 118,011 $ 21,825,537 *Sullivan J, Michael, 2009 Estimated Value of Service Reliability for Electric Utility Customers in the US, xxi

July 5, 2012 Storm Response (EPB) Courtesy of EPB of Chattanooga EPB of Chattanooga estimated that power outages resulted in an annual cost of $100 million to the community and installed automated fault isolation and service restoration technology (1200 automated feeder switches) 2011 Labor Day Storm (20% technology configured): 63,000 homes interrupted; however, 16,000 (25%) experienced no outage and 9,000 (7%) experienced a 2-second interruption Utility avoided 1,917,000 customer minutes of interruption July 2012 wind storm: EPB estimates that due to automated feeder switching with AMI they were able to avoid 500 truck rolls and reduce total restoration time by 1.5 days Represents $1.4 million in operational savings Courtesy of EPB of Chattanooga

Synchrophasor Technology for Transmission System Operations DOE and NERC/NASPI are working together closely with industry to enable wide area time-synchronized measurements that will enhance the reliability of the electric power grid through improved situational awareness and other applications April 2007 November 2012

Electrical Island during Hurricane Gustav Hurricane Gustav is second most destructive storm in 95-year history of Entergy’s utility system Hurricane Gustav made landfall on Sept 1, 2008 at 9:30 am (Category 2 hurricane) Customer outages from Gustav peaked at 964,000 second only to 1.1 million outages in 2005 during Hurricane Katrina (HG caused outage of 241 miles of transmission lines) 14 of transmission lines serving Baton Rouge – New Orleans area tripped out of service during the storm At approx 2:49 PM on September 1, 2008 the Baton Rouge – New Orleans island was created by the outage of the Gypsy to Fairview 230 kV transmission line, which is a 55-mile line running acrosss Lake Pontchartrain. Energy Phasor Measurement System has a total of 21 PMUs across 4 states. During the islanding event, the Waterford (WTFR) PMU was located in the island. The Mabelvale, AR PMU was choosen as the reference within the Eastern Interconnect Location of the PMUs proved vital in detecting and managing the island; able to minitor real-time changes Having PMUs GPS time synchronized and taking frequency measurements at 30 samples/second gave Entergy an advantage not possible with SCADA (which collects data once every 2 – 4 seconds and non-synchronized. When Waterford and Mabelvale PMUs were compared, the operators saw the frequency oscillations created from the birth of the island and were alerted of its creation. Island existed for 33 hours. At 2:49 PM on September 1, 2008 the Baton Rouge – New Orleans island was created due to outage of 230 kV transmission line Courtesy of Entergy

Monitoring Oscillations in the Island Governors of the three online generation units were set to address the potential for high frequency oscillations Governors of 3 online units in the island were in automatic mode prior to the formation of the island. Once the island formed, the elctrical strength of the island was much less than when connected to the grid and any variation in load demand could lead to high frequency oscillations. Oscillations could result when all 3 generators attempt to respond to a change in demand because of their pre-island-formation governor droop setting To prevent this scenario and stabilize the frequency in the island, the governors of the Waterford and Gypsy units were locked down and placed on manual mode; Ninemile put on automatic mode allowing it to swing to meet the varying load changes in the island. The PMU system warned of hunting among the three units when the governor of the gypsy facility was inadvertently switched from manual to automatic mode. PMUs warned dispatchers of the hunting condition and allowed control room personnel to make quick decisions. Hunting: Hunting is a phenomenon that can occur upon load changes in which the frequency or the voltage continues to rise above and fall below the desired value without reaching a steady-state value. It is caused by insufficient damping. Courtesy of Entergy

Reconnection to the Grid After 33 hours, the island was reconnected to the main grid Courtesy of Entergy

For More Information Contact: joseph.paladino@hq.doe.gov Websites: www.oe.energy.gov www.smartgrid.gov Reports: SGIG Progress Report (July 2012) Peak Demand Reductions – Initial Results (December 2012) AMI O&M Savings – Initial Results (December 2012) Reliability Improvements – Initial Results (December 2012) Voltage Optimization – Initial Results (December 2012) Economic Impact (April 2013) Customer Enrollment Patterns in Time-Based Rate Programs (June 2013) All reports are downloadable from: Please check out these web sites for more information on the U.S. DOE’s and the White House smart grid activities. Thanks for asking DOE to participate today. I look forward to your questions. http://www.smartgrid.gov/all/news/department_energy_releases_smart_grid_impact_reports 13