Taking the Measure of the around the World affiliateAdpe.indd 1 1/6/11 8:31 PM www.ieee-pes.org Special Issue: on the cover Taking the Measure of the Smart Grid around the World …a 2011 reprint journal from PES

Pictured on the front cover…Hong Kong. Pictured on the back cover…New York City features 3 Leader’s Corner A message from the President of PES, Alan C. Rotz

4 The Future of Power Transmission Technological Advances for Improved Performance Stanley H. Horowitz, Arun G. Phadke, and Bruce A. Renz

11 Getting Smart With a Clearer Vision of the Intelligent Grid, Control Emerges from Chaos Enrique Santacana, Gary Rackliffe, Le Tang, and Xiaoming Feng

19 A New Car, A New Grid The Electric Car is Back (with Help from New Batteries, a Smarter Grid, and Uncle Sam) Larry Dickerman and Jessica Harrison 26 Taking Demand Response to the Next Level 11 Leveraging the Experience of U.S. Utilities to Demonstrate the Smart Grid’s DR Potential Katherine Hamilton and Neel Gulhar

33 Demanding Standards Developing Uniformity in Wholesale Demand Response Communications to Enhance Industry Growth Scott Coe, Andrew Ott, and Donna Pratt

38 Get Smart Using Demand Response with Appliances to Cut Peak Energy Use, Drive Energy Conservation, Enable Sources and Reduce Greenhouse-Gas Emissions T. Joseph Lui, Warwick Stirling, and Henry O. Marcy

51 Engineering the Future A Collaborative Effort to Strengthen the U.S. Power and Energy Workforce 38 Wanda Reder, Anjan Bose, Alex Flueck, Mark Lauby, Dagmir Niebur, Ann Randazzo, Dennis Ray, Gregory Reed, Peter Sauer, and Frank Wayno ieee power & energy magazine 1 m a g a z i n e

Editor in Chief IEEE Periodicals/Magazines Department Melvin I. Olken 445 Hoes Lane, Piscataway, NJ 08854 USA 245 East 19th Street #20K +1 732 562 3950, fax +1 732 981 1855 New York, NY 10003-2665 USA www.ieee.org/magazines +1 212 982 8286 (phone fax) Geraldine Krolin-Taylor, Senior Managing Editor [email protected] Janet Dudar, Senior Art Director Gail A. Schnitzer, Assistant Art Director Associate Editors Theresa L. Smith, Production Coordinator Gerald B. Sheblé, Business Scene Peter M. Tuohy, Production Director Felicia Spagnoli, Advertising Production Manager Carl L. Sulzberger, History Dawn Melley, Editorial Director Editorial Board Fran Zappulla, Staff Director, IEEE Publishing Operations IEEE prohibits discrimination, harassment, and bullying. For more information, S. Massoud Amin, L. Goel, A.P. Hanson, visit http://www.ieee.org/web/aboutus/whatis/policies/p9-26.html. N. Hatziargyriou, M.I. Henderson, S.H. Horowitz, P. Kundur, R. Masiello, IEEE Power & Energy Magazine K.M. Matsuda, A.P.S. Meliopoulos, IEEE Power & Energy Magazine (ISSN 1540-7977) is published bimonthly by the Institute of Electrical and Electron- M.I. Olken, M. O’Malley, A.G. Phadke, ics Engineers, Inc. Headquarters: 3 Park Avenue, 17th Floor, New York, NY 10016-5997 USA. Responsibility for the contents C.E. Root, H. Rudnick, P.W. Sauer, rests upon the authors and not upon the IEEE, the Society, or its members. IEEE Operations Center (for orders, subscrip- tions, address changes): 445 Hoes Lane, Piscataway, NJ 08854 USA. Telephone: +1 732 981 0060, +1 800 678 4333. M. Shahidehpour, B.R. Shperling, Individual copies: IEEE members $20.00 (first copy only), nonmembers $66.00 per copy. Subscription Rates: Society V.C. Sulzberger, S.S. Venkata, members included with membership dues. Subscription rates available upon request. Copyright and reprint permissions: B.F. Wollenberg Abstracting is permitted with credit to the source. Libraries are permitted to photocopy beyond the limits of U.S. Copyright law for the private use of patrons 1) those post-1977 articles that carry a code at the bottom of the first page, provided the per-copy fee indicated in the code is paid through the Copyright Clearance Center, 222 Rosewood Drive, Danvers, MA Advertising 01923 USA; 2) pre-1978 articles without fee. For other copying, reprint, or republication permission, write Copyrights and Barry LeCerf Permissions Department, IEEE Operations Center, 445 Hoes Lane, Piscataway, NJ 08854 USA. Copyright © 2011 by the Bullseye International Group, Inc. Institute of Electrical and Electronics Engineers, Inc. All rights reserved. Periodicals postage paid at New York, NY, and at additional mailing offices. Postmaster: Send address changes to IEEE Power & Energy Magazine, IEEE Operations +1 913 663 1112, fax +1 913 663 1119 Center, 445 Hoes Lane, Piscataway, NJ 08854 USA. Canadian GST #125634188 [email protected] Printed in U.S.A. IEEE POWER & ENERGY SOCIETY (PES) The IEEE Power & Energy Society is an organization of IEEE members whose principal interest is the advancement of the science and practice of electric power generation, transmission, distribution, and utilization. All members of the IEEE are eligible for membership in the society. Mission Statement: To be the leading provider of scientific and engineering information on electric power and energy for the betterment of society, and the preferred professional development source for our members. Officers PES Executive Director Technical Council A.C. Rotz, President Patrick Ryan, +1 732 465 6618, J. Nelson, Vice Chair N.N. Schulz, President-Elect fax +1 732 562 3881, e-mail [email protected] S.S. Venkata, Secretary M. Selak, Vice President, Chapters D. Novosel, Vice President, Technical Activities Standing Committee Chairs Technical Committee Chairs P.W. Sauer, Vice President, Education S. Ellis, Awards & Recognition O. Malik, Electric Machinery M. Shahidehpour, Vice President, Publications D. Novosel, Constitution & Bylaws M. Begovic, Emerging Technologies W. Rosehart, Vice President, Meetings M.M. Begovic, Finance & Audit L. Wozniak, Energy Development H. Louie, Vice President, W.K. Reder, Nominations & Appointments & Power Generation Membership & Image K. Butler-Purry, Power Engineering Education R.I. Mosier, Insulated Conductors S. Rahman, Vice President, New Initiatives/ D. Von Dollen, Intelligent Grid Chapter Representatives P. Bishop, Marine Systems Outreach B. Allaf, C. Chamochumbi, Y. Chen, M.M. Begovic, Treasurer J. MacDonald, Nuclear Power Engineering D.J. Doke, M. Eremia, J. Fleemans, J. McConnach, Policy Development C.E. Root, Secretary L. Goel, A. Hintze, Z.F. Hussien, W.K. Reder, Past-President K. Tomsovic, Power System Analysis, S. Khan, M. Kiani, I. Kuzle, N. Logic, Computing, & Economics Division VII Director R. Mazzatto, P. Naidoo, B. Presat, F. Cleveland, Power System Communications E.A. Tejera N. Segoshi, K. Shahidul, E. Tobin, S. Vega N. Hatziargyriou, Power System Dynamic Performance Division Director Elect Chapter Committee Chairs R. Arseneau, Power System C. Warren J. Hollman, Membership Data Coordinator Instrumentation & Measurements Region Representatives N. Marium, Chapter/Section Relations R.J. Kafka, Power System Operations Y. Chen, F. Lambert, A. Lazarski, M. Armstrong, Electronic Communications M.L. Chan, Power System Planning M. Nissen, K. Sen, E. Carlsen, Awards & Resources & Implementation E. Uzunovic, United States K. Taylor, Distinguished Lecturer Program M.P. Sanders, Power System Relaying M. Wactor, Standards H. Turanli, Canada Membership & Image C. Vournas, Europe, Middle East, & Africa J. LaMarca, Stationary Battery Committee Chairs H.J. Koch, Substations J.C. Miguez, Latin America A. St Legec, GOLD Coordinator L. Goel, Asia & Pacific F. Waterer, Surge Protective Devices A. Bonthron, Membership Development B. Long, Switchgear Governing Board L. Fan, Membership Development T. Prevost, Transformers Members-at-Large Open, Web Site Development T.E. Grebe, Transmission & Distribution E. Gunther, M. Jensen, T. Prevost M.A.S. Althani, PES WIE Liaison R.J. Piwko, Wind Power Digital Object Identifi er 10.1109/MPE.2010.939789

2 IEEE power & energy magazine Alan C. Rotz, President, IEEE PES

greetings from the IEEE Power & Energy Society (PES) leader’s corner leader’s

January 15, 2011 take advantage of efficiencies that SG TO: Recipients of this 2011 IEEE PES can offer: Smart Grid Reprint Journal: ✔ Why standards are a critical part Greetings from the IEEE Power & Energy of any SG Solution Society (PES)! ✔ how smart appliances intelligently integrated into the SG can provide consumer control as well as imme- ThE 2010 EdITIon of ThIS SmarT diate efficiency benefits Grid reprint Journal was so well-received ✔ a vision of what a Smart Trans- that PES decided to create a 2011 Edi- mission Grid might look like in tion. The enclosed articles were all have a common look. Companies will the near future published in PES’s Power & Energy first design their SG implementations as a bonus, we’ve included an article magazine during 2010. The articles are to deliver the results they individually by former PES President Wanda reder, re-packaged into this special issue as a see as beneficial, be it reliability, cost et al, that discusses the demographics Thandy reference on various perspec- savings, security, or increased services. of engineers in the power & energy in- tives on Smart Grid (SG). as most of however, the equipment used must be dustry and what one group has been you know, the Grid has been “smart” capable of interoperability with all other working on to address those shortages. for many years. But in the years to equipment that they or other utilities If you are not an IEEE and PES come, it will be getting smarter. Pilot will use. That’s where standards come member, we invite you to join us in this projects have sprung up in many loca- in. PES and other IEEE Societies have challenging and exciting effort to lead tions at all voltages that will hopefully been working hard to update and create the power and energy industries to new provide successful models for full-scale those standards to make sure this horizons. Your ability to network with implementations. PES and other IEEE interoperability goal is met. the brightest minds in the industry will Societies have been working on stan- a compendium of this size will never pay back your investment many times dards and technology for many years be able to deliver all the information over. In addition, if you’re ready to that allow the grid to operate at the that every individual is looking for on learn more about Smart Grid, we invite high levels of security and reliability SG. however, we’ve tried to include a you to log onto the IEEE Smart Grid that we enjoy today. many of these diversity of articles on relevant current Web Portal at the following link: standards have been and will continue and future topics that must be considered (http://smartgrid.ieee.org/). This site to be updated as new technologies for when implementing a SG solution. is intended to be your key web page for the SG evolve. here’s a sampling of what you’ll find in everything about Smart Grid. We Clearly, SG is a term that can apply this edition: recommend you save it as a “favorite” at any level of the electrical network a discussion on how the Generation, and return often to check progress. from the generation plants to the T&d and end users must all work With warmest regards, customer premises beyond the meter, together, and the layers of technology as well as to every level in between. that must be applied to make that happen. however, even at the same voltage Why demand response programs levels of the power system, individual are a critical element for customers to alan C. rotz SG pilots and implementations may not manage their electric consumption and IEEE PES President

ieee power & energy magazine 3 Reprinted from March/April 2010 issue of Power & Energy magazine

The Future of Power ©Brand x pictures Transmission

The elecTric power sysTem is on The verge Technological of significant transformation. For the past five years or so, work has been under way to conceptualize the shape of a Advances 21st-century grid that exploits the huge progress that has been made in digital technology and advanced materials. for Improved The national energy Technology laboratory (neTl) has identified five foundational key technology areas (KTAs), as Performance shown in Figure 1. Foremost among these KTAs will be integrated communi- cations. The communications requirements for transmission enhancement are clear. Broadband, secure, low-latency chan- Tnels connecting transmission stations to each other and to con- trol centers will enable advances in each of the other KTAs. ✔✔ sensing and measurements will include phasor mea- surement data streaming over high-speed channels. By Stanley H. Horowitz, ✔✔ Advanced components, such as all forms of flexible Arun G. Phadke, ac transmission system (FAcTs) devices, hvDc, and new storage technologies will respond to control sig- and Bruce A. Renz nals sent to address perturbations occurring in mil- Digital Object Identifier 10.1109/MPE.2009.935554 liseconds.

4 ieee power & energy magazine 1540-7977/10/$26.00©20101540-7977/10/$26.00©2010 IEEE IEEE march/april 2010 ✔✔ Advanced control (and protection) methods will in- clude differential line relaying, adaptive settings, and Advanced various system integrity protection schemes that rely Control on low-latency communications. Methods ✔✔ Improved interfaces and decision support will uti- lize instantaneous measurements from phasor mea- Sensing and Integrated Decision Support surement units (PMUs) and other sources to drive Measurements Communications fast simulations and advanced visualization tools that can help system operators assess dynamic chal- lenges to system stability. Advanced Each of these elements will be applied to the moderniza- Components tion of the grid, at both the distribution level and the transmis- sion level. Because it is clearly less advanced, distribution is figure 1. NETL’s five key technology areas. receiving most of the initial focus. This is dramatically illus- trated by the American Recovery and Reinvestment Act’s Smart Grid Investment Grants (SGIGs), announced in Octo- mission system must be based on a study of the past con- ber. Of the $3.4 billion awarded to 100 proposers (of the more sidering its successes and failures, on knowledge of the than 400 that applied), only $148 million went to transmission existing system and all of its component disciplines, and applications; most of the rest was for distribution projects. on a thorough understanding of the latest technologies and While the changes to distribution will be revolutionary, their possible applications. The electrical power system, transmission will change in an evolutionary manner. Dis- and in particular its transmission and distribution network, tributed generation and storage, demand response, advanced is a vital and integral part of today’s society. Because it metering infrastructure (SGIGs will fund the deployment of is essential to all our endeavors, we must be prepared to 18 million smart meters), distribution automation, two-way integrate new, exciting, and highly innovative concepts to power flow, and differentiated power quality together rep- guarantee that it performs reliably, safely, economically, resent a sea change in distribution design that will require and cleanly. enormous financial and intellectual capital. Although not unique in world events involving power The role of transmission will not be diminished, how- systems, two widely known outages in the United States and ever, by this new distribution paradigm. Large central power Canada serve as examples of the history, analysis, and reme- plants will continue to serve as our bulk power source, and dies for blackouts and can provide a basis for future actions. many new ones will be fueled by renewable resources that Widely publicized, the blackouts of 9 November 1965 in would today be out of reach of the transmission grid. New the northeastern United States and 23 August 2003 in the lines will be built to connect these new plants, and new northeastern United States and Canada are typical events methods will be employed to accommodate their very dif- that can help shape our planning and operating efforts for ferent performance characteristics. Addressing the result- the future. ing greater variability of supply will be the job of the five In 1965 we learned that cooperation and interaction KTAs listed above. As KTA technology speeds increase, between utilities were essential. In response to the blackout, control of transmission will advance from quasi-steady- utilities established the National Electric Reliability Cor- state to dynamic. poration (NERC) in 1968, which began distributing recom- The traditional communications technologies capable of mendations and information. These communications formed supporting these strict requirements are fiber optics (e.g., the basis for more reliable and secure planning, operating, optical ground wire) and microwave. Recently a third can- and protective activities. The decisions of the newly formed didate has appeared on the scene. Research funded by the NERC were, however, only recommendations. Deficiencies U.S. Department of Energy (DOE) and American Electrical due to limitations in transmission planning, operations, and Power in conjunction with a small Massachusetts smart-grid protection were recognized, and steps were taken to correct communications company, Amperion, has demonstrated the them. Transmission systems were strengthened considerably viability of broadband over power line (BPL) for application by the construction of 345-kV, 500-kV, and 765-kV lines. on transmission lines. Currently, a five-mile, 69-kV line is System planning studies were made cooperatively; operat- operating at megabit-per-second data rates with latency of ing parameters and system problems were studied jointly. less than 10 ms. The next step will be to extend this high- Underfrequency load shedding became universal, with spe- voltage BPL technology to 138 kV. cific settings arrived at by agreement between utilities, and loss-of-field relaying was recognized as a system phenom- How We Got Here enon and studied accordingly. In 551 B.C., Confucius wrote, “Study the past if you would In 2003 another blackout of similar proportions affected know the future.” The future of the electrical power trans- the northeastern United States and parts of Canada. The march/april 2010 ieee power & energy magazine 5 causes of that event included not recognizing load and sta- electricity markets, the value of these various electronics- bility restrictions and, unfortunately, human error, which based power devices will only grow. suggested that improved systemwide monitoring, alarms, In addition to FACTS, bulk storage, and CLDs, vari- and power system state estimation programs would be ous new aspects of the distribution system such as demand useful and should be instituted. The ability of a distance response, , plug-in hybrid electric relay to differentiate between faults and load, particularly vehicles (PHEVs), and other forms of distributed storage can when the system is stressed, has become a major concern. be centrally coordinated and integrated to function as a “vir- NERC requires that this condition be included in relay set- tual power system” that supports the transmission system in ting studies. times of stress. In 1920, Congress founded the Federal Power Com- mission (FPC) to coordinate hydroelectric power devel- Advanced Protection opment. Fifty-seven years later, in response to the energy Recognizing that protection of specific equipment and crisis, the DOE was formed. The DOE included the FPC, localized systems is inadequate in the face of systemwide renamed the Federal Energy Regulatory Commission stress, in 1966 a joint IEEE/CIGRE questionnaire was (FERC), whose mandate was primarily to conduct hear- circulated. The results indicated that protection schemes ings and approve price control and related topics, including had to encompass wider areas of the transmission system. electric practices on bulk transmission systems. After the This effort required communication and control center 2003 event, FERC also became a regulatory instrument, involvement. The effort was termed special protection sys- reviewing transmission line improvements and rights-of- tems (SPS). The primary application of SPS at that time way. FERC review and approval, as with NERC, has now was for limited system events such as underfrequency and become mandatory. The actions of FERC and NERC will, undervoltage, with some advanced generation controls. in the future, be major components of system decisions As system stress becomes a more common concern, the and practices. application of SPS takes on added importance and in fact becomes an important tool for protecting the grid against Technology’s Role Going Forward wide-area contingencies. With the preceding as background, we can now review in The SPS concept is no longer considered “special” and greater detail some of the transmission enhancements that is now commonly referred to as system integrity protection will be part of the 21st-century transmission system. systems (SIPS), remedial action systems (RAS), or wide- area protection and control (WAPC). These schemes are Advanced Control intended to address widespread power system constraints or It is axiomatic that the fundamental basis for the reliable to be invoked when such constraints could occur as a result performance of the transmission system has to be the system of increased transfer limits. The Power System Relaying itself. The primary components, system configuration, line Committee of the IEEE initiated a recent survey on power specifications, and design of high-voltage equipment must system integrity protective schemes that was distributed be consistent with the mission of the power system, i.e., to worldwide with cooperation from CIGRE, NERC, IEE, and deliver electric energy safely, reliably, economically, and in a other utility organizations. The survey revealed very wide- timely fashion. Furthermore, high-voltage, electronic-based spread application, with more than 100 schemes of various power equipment such as bulk storage systems (e.g., flow complexity and purpose. Emerging technologies in high- batteries), FACTS devices (including unified power flow speed communication, wide-area measurement, and phasor controllers, static var compensators, and static synchronous measurement are all employed and will be vital components compensators), and current-limiting devices (CLDs), which of the transmission system in the future. are based on high-temperature superconductivity, are now One of the most exciting features of the transmission sys- or will soon be available. Coupled with sophisticated com- tem of the future involves power system protection. This is munications and computing tools, these devices make the due in large measure to the advantages of digital technology transmission system much more accommodating of varia- for relays, communication, and operation. Relays now have tions in load and/or voltage. the ability to perform previously unimaginable functions, Of these advanced control devices, FACTS represents made feasible by evaluating operating and fault parameters the most mature technology. It is in somewhat limited use and coupling this data with high-speed communication and at present but has the potential to be an increasingly impor- computer-driven applications within the power system con- tant element in the future. FACTS can provide control of trol center. With the ever-increasing restrictions on trans- ac transmission system parameters and thus increase power mission line and generator construction and siting and the transfer capability and improve voltage regulation. Changes decreasing difference between normal and abnormal opera- in generation and load patterns may make such flexibility tion, loading, and stability, the margins between the relay- extremely desirable. With the increased penetration of central ing reliability concepts of dependability and security are renewable sources and with the continued variability of becoming blurred. Consequently, the criteria of traditional

6 ieee power & energy magazine march/april 2010 While the changes to distribution will be revolutionary, transmission will change in an evolutionary manner.

protection and control are being challenged. The hallmark fer this information to analyzers have made previously used of relays is the tradeoff between dependability (the ability to oscillography and sequence-of-events recorders obsolete. always trip when required) and security (the characteristic Replacing these devices will result in very significant sav- of never tripping when not required). ings in both hardware and installation costs. AEP, in con- Traditionally, relays and relay schemes have been designed junction with Schweitzer Engineering and Tarigma Corp., to be dependable. Losing a transmission line element must has embarked on a revolutionary program that lets selected be tolerated, whether the loss is for an actual fault or for an centers receive data from critical substations that will com- inadvertent or incorrect trip. When the system is stressed, bine, display, and analyze fault data to a degree and in a however, an incorrect trip is not allowable. With the system time frame heretofore not possible. Combining the current, stressed, losing another element could be the final step in voltage, communication signals, and breaker performance bringing down the entire network. With digital logic and from several stations on one record that can be analyzed at operations, it is possible to reorder protection priorities and several control and engineering centers permits operations require additional inputs before allowing a trip. This can be to be verified and personnel to be alerted to potential prob- done with appropriate communication from a central center lems. A vital by-product of this advanced monitoring is the advising the relays. fact that it allows NERC requirements for monitoring and Probably one of the most difficult decisions for a relay is analysis to be met. to distinguish between heavy loads and faults. Heretofore, Perhaps even more exciting is the possibility of predicting relays simply relied on the impedance measurement, with the instability of a power swing. Modern protection theory settings determined by off-line load studies using conditions knows how to detect the swing using zones of stability and based on experience. As in the two blackouts mentioned instability. The problem is how to set the zones. With accu- above, this criterion was not adequate for unusual system rate synchronized phasor measurements from several buses, conditions that were not previously considered probable. the goal of real-time instability protection seems achievable. Digital relays can now establish such parameters as power Out-of-step relays could then establish blocking or tripping factor or voltage and remove the measured impedance from functions at the appropriate stations. the tripping logic. The role of underfrequency load shedding has already The bête noir of protection has traditionally been the been discussed. Future schemes, however, could use real- multiterminal line. The current to the fault and the voltage time measurements at system interconnection boundaries, at the fault defines the fault location. A relay designed to compute a dynamic area control error, and limit any poten- protect the system for this fault, however, sees only the cur- tial widespread underfrequency by splitting the system. rent and voltage at that relay’s specific location. The advent Computer relays, if not already in universal use, will be of high-speed communication and digital logic remedies this in the near future. This will let utilities protect, monitor, and condition and allows all involved relays to receive the appro- analyze system and equipment performance in ways and to a priate fault currents and voltages. degree not possible before. The increasing popularity of transmission line differen- tial relaying also provides both dependability and security Synchronized Phasor Measurements for faults in a multiterminal configuration. Although primar- It has been recognized in recent years that synchronized ily a current-measuring relay, the digital construction allows phasor measurements are exceedingly versatile tools of far more protection, monitoring, and recording functions. modern power system protection, monitoring, and control. Future applications will be available to accomplish the fea- Future power systems are going to depend on making use tures mentioned above and in ways not yet implemented or of these measurements to an ever-increasing extent. The even thought of. principal function of these systems is to measure positive One of the earliest advantages of the computer relay sequence voltages and currents with a precise time stamp is its ability to monitor itself and either repair, replace, or (to within a microsecond) of the instant when the measure- report the problem. This feature is sure to be a major fea- ment was made. The time stamps are directly traceable to ture in future transmission line protection. In addition, the the Coordinated Universal Time (UTC) standard and are information stored in each relay during both normal and achieved by using Global Positioning System (GPS) trans- abnormal conditions and the ability to analyze and trans- missions for synchronization. Many PMUs also provide march/april 2010 ieee power & energy magazine 7 other measurements, such as individual phase voltages and grid and its ability to handle contingency conditions that currents, harmonics, local frequency, and rate of change of may occur in the immediate future. This was a big step for- frequency. These measurements can be obtained as often ward in intelligent operation of the power grid. The limita- as once per power frequency cycle, although for a number tions of this technology (such as nonsimultaneity of system of applications a slower measurement rate may be prefer- measurements across the network) were rooted in the tech- able. In well-designed systems, measurement latency (i.e., nology of that day. The fact that the data from a dynamically the delay between when the measurement is made and when changing power system was not obtained simultaneously it becomes available for use) can be limited to fewer than over a significant time span meant that the estimated state 50 ms. The performance requirements of the PMUs are was an approximation of the actual system state. Conse- embodied in the IEEE synchrophasor standard (C37.118). A quently, the system state and its response to contingencies measurement system that incorporates PMUs deployed over could only be reasonably accurate when the power system large portions of the power system has come to be known as was in a quasi-steady state. Indeed, when the power system a wide-area measurement system (WAMS), and a power sys- was undergoing significant changes due to evolving events, tem protection, monitoring, and control application that uti- the state estimator could not always be counted on to con- lizes these measurements is often referred to as a wide-area verge to a usable solution. measurement protection and control system (WAMPACS). The advent of wide-area measurements using GPS- synchronized PMUs led to a paradigm shift in the state Automatic Calibration of estimation process. With this technology, the capability of Instrument Transformers directly measuring the state of the power system has become It is well known that current and voltage transformers a reality. PMUs measure positive sequence voltages at net- used on high-voltage networks have ratio and phase-angle work buses and positive sequence currents in transmission errors that affect the accuracy of the measurements made lines and transformers. Since the state of the power system on the secondary of these transformers. Capacitive volt- is defined as a collection of positive sequence voltages at age transformers are known to have errors that change with all network buses, it is clear that with sufficient numbers of ambient conditions as well as with the age of the capacitor PMU installations in the system one can measure the sys- elements. Inductive instrument transformers have errors that tem state directly: no estimation is necessary. In fact, the change when their secondary loading (burden) is manually transmission line currents provide a direct estimate of volt- changed. The PMU offers a unique opportunity for cali- age at a remote bus in terms of the voltage at one end. It is bration of the instrument transformers in real time and as therefore not necessary to install PMUs at all system buses. often as necessary. In simple terms, the technique is based It has been found that by installing PMUs at about one-third on having some buses where a precise voltage transformer of system buses with voltage and current measurements, it (with known calibration) is available and where a PMU is is possible to determine the complete system state vector. placed. Potential transformers used for revenue metering are Feeding this information into the appropriate computers an example of such a voltage source. Using measurements by provides the information necessary for the adaptive protec- the PMU at this location, the calibration at the remote end tive function described above. Of course, a larger number of a feeder connected to this bus can be found. This calibra- of PMUs provides redundancy of measurements, which is tion is not affected by current transformer (CT) errors when always a desirable feature of estimation processes. the system loading is light. It is thus possible to calibrate all voltage transformers using current measurements at light Complete and Incomplete Observability system load. Using the voltage transformer calibration thus In order to achieve a state estimate in the traditional way, obtained and additional measurements during heavy system i.e., by using unsynchronized supervisory control and data load, the current transformers can be calibrated. In practice, acquisition (SCADA) measurements, a complete network it has been found (in simulated case studies) that by combin- tree must be measured. With PMUs, however, it is sufficient ing several light and heavy load measurement sets a very to measure isolated parts of the network, which provides accurate estimate of all the current and voltage transformers islands of observable networks. This is possible since all can be obtained. Although a single accurate voltage source phasors are synchronized to the same instant in time. The is sufficient in principle, having a number of them scattered process has been described as PMU placement for incom- throughout the network provides a more secure calibration. plete observability. The remaining network buses can be estimated from the observed islands using approximation Precise State Measurements and Estimates techniques. This is, of course, not as accurate as providing a State estimation of power systems using real-time measure- sufficient number of PMUs in the first place. But it has been ments of active and reactive power flows in the network shown that combining incomplete observations with such an (supplemented with a few other measurements) was intro- approximation technique to estimate the unobserved parts duced in the late 1960s to improve the awareness among provides surprisingly useful results. Incomplete observabil- power system operators of the prevailing state of the power ity estimators are a natural step in the progression towards

8 ieee power & energy magazine march/april 2010 A common problem faced by interconnected power systems is that various parts of the system may be under different control centers, with each part having its own state estimator.

complete observability and will be a feature in future trans- all partial control centers be determined or 2) an alternative mission systems. must be found to modify the results of individual state esti- Figure 2 illustrates the principle of complete and incom- mates to put all states on a common reference. Option 1 is plete observability. In Figure 2(a), PMUs are placed at buses cumbersome and wasteful of computational effort. Option identified by dark circles. By making use of the current mea- 2 becomes exceedingly simple with PMUs. At the simplest surements and the network impedance data, it is possible to level, one can visualize putting a PMU at each of the ref- calculate the voltages at the buses identified by light blue erence buses, thus obtaining the phase-angle relationships circles. In this case, complete observability is achieved with between all partial estimates. These phase-angle corrections two PMUs. Figure 2(b) illustrates the use of fewer PMUs may then be used to form a combined state estimate for the than would be necessary for complete observability. Even entire interconnected network on a single reference. It has with current measurements, it is not possible to determine been found in practice that the placement of a few PMUs the voltages at the buses identified by the red circles. These in each partial system (rather than just one at the reference buses form islands of incomplete observability. As men- bus) leads to greater security and optimal performance. This tioned earlier, these bus voltages can be estimated fairly principle is illustrated in Figure 3. Systems 1 and 2 are con- accurately using voltages at surrounding buses. nected by tie lines and have state estimates S1 and S2 that are obtained independently, each with its own reference bus. State Estimates of Interconnected Systems With the use of PMU data from optimally selected buses A common problem faced by interconnected power systems (shown in red), it is possible to determine the angle differ- is that various parts of the system may be under different ence between the two references and obtain a single state control centers, with each part having its own state estimator. estimate for the interconnected system. This, of course, implies that each partial state estimate has its own reference bus. To perform studies such as contingency Intelligent Visualization Techniques analysis on the interconnected power system, it is necessary The traditional visualization techniques used in energy man- to have a single state estimate for the entire network. This agement system (EMS) centers focus on showing network bus requires either that 1) a new system state using data from voltages and line flows, along with any constraint violations that may exist. It is, of course, possible to reproduce such displays using WAMS technology. Dynamic loading limits of transmission lines have been estimated with WAMS, and it would be relatively simple to show prevailing loading con- ditions and their proximity to the dynamic loading limits. PMU Many PMUs offer the possibility of measuring system unbal- Indirect ances. It would then be possible to display unbalance cur- rents to determine their sources and mitigation techniques to correct the unbalance.

(a)

Reference Reference PMU Indirectly Observed Unobserved

Traditional SE Result: E1 Traditional SE Result: E2

(b) Optimal Placement of PMUs

figure 2. (a) Complete observability. (b) Incomplete figure 3. Connecting adjacent state estimates with observability. phasor data. march/april 2010 ieee power & energy magazine 9 Conclusion Modernizing the U.S. power grid has become a national pri- ority. Unprecedented levels of governmental funding have been committed in order to achieve this goal. The initial focus has been on the fundamental transformation of the distribution system. This is in itself a huge technical chal- lenge that will be measured not in years but in decades. The end result is expected to be higher efficiency, reduced envi- ronmental impact, improved reliability, and lower exposure to terrorism. The revolution in distribution must be accompanied by the continued evolution of the transmission system. Events like the 2003 blackout—more the result of human shortcom- ings than technological breakdowns—can be eliminated by exploiting the huge progress made in recent years in the digital and material sciences. Other industries have already harvested these opportunities; now it is our turn. figure 4. U.S. Phasor contour map. Technological development is an engineering challenge. This nation has time and again demonstrated its ability With direct measurement of synchronized phasors, to meet such challenges whenever they have been clearly many more display options become possible. For example, a focused. But there is another challenge that may actually geographical display with phase angles at all network buses be more difficult. It is to find the political alignment that is shown at the physical location of buses—and perhaps fitted needed to accept the vision and move forward aggressively. with a surface in order to provide a hilly contour—would For transmission, that means recognizing that new lines, not immediately show the distribution of positive sequence volt- just better lines, will be needed. It is simply not acceptable age phase angles. to wait ten or more years for a new line to move from con- Figure 4 shows such a visualization of a hypothetical net- cept to reality. Unlike many other parts of the world, the work state for the entire United States. The map colors iden- United States has allowed fragmented responsibility for tify the magnitude and sign of the positive sequence voltage transmission additions to slow the process to an unaccept- phase angle with respect to a center of angle reference. The able extent. lower plot is a footprint of equiangle loci from the map. Since With the intense focus now on energy in general and the positive sequence voltage phase-angle profile of a net- electricity in particular, it should be possible to overcome work conveys a great deal of information regarding its power both the technical and the political obstacles and to reestab- flow and loading conditions, such visualizations can instantly lish U.S. leadership in this vital arena. Doing so is a matter show the quality of the prevailing system state and its dis- of huge national significance that will affect the lifestyle of tance from a normal state. High-speed dynamic phenomena all Americans in this new century. can be represented by animations of such visualizations. Such a display would instantly show the general dispo- For Further Reading sition of generation surplus and load surplus areas. Such a V. Madani and D. Novosel, “Getting a grip on the grid,” picture can be updated at scan rates of a few cycles, leading IEEE Spectr., pp. 42–47, Dec. 2005. to visualization of dynamic conditions on the network. If P. Anderson and B. K. LeReverend, “Industry experience thresholds for phase-angle differences between key buses with special protection schemes,” IEEE Trans. Power Syst., have been established for secure operation of the network, vol. 2, no.3 , pp. 1166–1179, Aug. 1996. then violation of those thresholds could lead to important “Global Industry Experiences with System Integrity Pro- alarms for the operator. Similarly, when islands are formed tection Schemes,” Survey of Industry Practices, IEEE Power following a catastrophic event, the boundaries of those System Relaying Committee, submitted for publication. islands could be displayed for the operator. Several protec- tion and control principles are being developed to make Biographies use of wide-area measurements provided by PMUs. Adap- Stanley H. Horowitz is a former consulting electrical engi- tive relaying decisions made in this manner could also be neer at AEP and former editor-in-chief of IEEE Computer displayed for the use of protection and control engineers. Applications in Power magazine. The technology of visualization using WAMS schemes is Arun G. Phadke is the University Distinguished Profes- in its infancy. As we gain greater experience with these sor Emeritus at Virginia Tech. systems, more interesting display ideas will undoubtedly Bruce A. Renz is president of Renz Consulting, LLC. be forthcoming. p&e

10 ieee power & energy magazine march/april 2010 Reprinted from March/April 2010 issue of Power & Energy magazine Getting Smart

With a Clearer Vision of the Intelligent Grid, Control Emerges from Chaos

It’s hard not to notIce In natIonaL news and professional conferences of the last few years all the talks and activities in the electric power industry about smart grids. “smart grid” and similar phrases (“intelligent grid,” “modern grid,” “future grid,” and so on) have all been used to describe a “digitized” and intelligent version of the present-day power grid. although there is some debate on what specifically constitutes a smart grid, a consensus is forming regarding its general attributes. Ithe following attributes of a smart grid are commonly cited in the United states: ✔✔ It is self-healing (from power disturbance events). ✔✔ It enables active participation by consum- ers in demand response. ✔✔ It operates resiliently against both physical and cyber attacks. ✔✔ It provides quality power that meets 21st- century needs. ✔✔ It accommodates all generation and storage options.

By Enrique Santacana, Gary Rackliffe, ©master series Le Tang, and

Digital Object Identifier 10.1109/MPE.2009.935557 Xiaoming Feng march/april 2010 1540-7977/10/$26.00©2010 IEEE ieee power & energy magazine 11 ✔✔ It enables new products, services, and markets. (transmission and distribution networks); and the end cus- ✔✔ It optimizes asset utilization and operating efficiency. tomers (residents, commercial buildings, industrial installa- The smartness of the smart grid lies in the decision intelligence In Europe, a smart grid is described, according to a recent tions, and others). layer, all the computer programs that run in relays, IEDs, substation European Commission report, as one that is The electric grid is unique in that electrical supply and ✔✔ flexible as it fulfills customers’ needs while respond- demand must remain tightly balanced at all times, since automation systems, control centers, and enterprise back offices. ing to the changes and challenges ahead for most of the history of the electric grid there has been ✔✔ accessible as it grants connection access to all net- no commercial solution for large-scale storage of electric- work users, particularly renewable power sources and ity to compensate for any excess or shortfall in power. In high-efficiency local generation with no or low carbon the past, this balancing act was performed by the vertically The scope of the smart grid extends over all the intercon- driver. Most of us consider cruise control a smart function of emissions integrated utilities that controlled both the generation and nected electric power systems, from centralized bulk gen- cars. Beyond the basic speed control, high-end automobiles ✔✔ reliable as it assures and improves security and quality the delivery systems. eration to distributed generation (DG), from high-voltage feature collision avoidance capability, using adaptive control of supply, consistent with the demands of the digital Power grids in the industrialized countries are aging and transmission systems to low-voltage distribution systems, and radar technologies. age, with excellent resilience in the face of hazards being stressed by operational scenarios and challenges never from utility control centers to end-user home-area networks, Figure 3 is a schematic diagram of a basic cruise control and uncertainties envisioned when the majority of them were developed many from bulk power markets to demand response service pro- system. Signals for vehicle speed, driver commands (set speed, ✔✔ economical as it provides best value through innova- decades ago. The main challenges are summarized below. viders, and from traditional energy resources to distributed increase or reduce speed, and so on), clutch and brake pedal posi- tion, efficient energy management, and “level playing ✔✔ Deregulation unleashed unprecedented energy trad- and renewable generation and storage, as shown in Figure 1. tions, and fuel injection throttle positions are fed to an onboard field” competition and regulation. ing across regional power grids, presenting power The transition from the present grid to a smart grid and cruise-control computer. There control programs work on the China is also developing the smart grid concept. “The flow scenarios and uncertainties the system was not the key differences between the two can be illustrated by input data continuously and, based on control theory algorithms, term ‘smart grid’ refers to an electricity transmission and dis- designed to handle. Figure 2. One can see there is tribution system that incorporates elements of traditional and ✔✔ The increasing penetration of renewable energy in the a fundamental shift in the de- cutting-edge power engineering, sophisticated sensing and system further increases the uncertainty in supply and sign and operational paradigm monitoring technology, information technology, and commu- at the same time adds stress to the existing infrastruc- of the grid: from central to dis- nications to provide better grid performance and to support a ture due to the remoteness of the geographic locations tributed resources, from predict- wide range of additional services to consumers. A smart grid where the power is generated. able power flow directions to is not defined by what technologies it incorporates but rather ✔✔ Our digital society depends on and demands a power unpredictable directions, from a by what it can do,” according to the nonprofit Joint US-China supply of high quality and high availability. passive grid to an active grid. In Cooperation on Clean Energy. ✔✔ The threat of terrorist attacks on either the physical or short, the grid will be more dy- Such high-level characterization of the smart grid, while the cyber assets of the power grid introduces further namic in its configuration and its helpful at the strategic level, leaves plenty of room for confu- uncertainty. operational condition, which will sion and very different interpretations on the part of laymen ✔✔ There is an acute need to achieve sustainable growth present many opportunities for and professionals alike, due to a lack of specifics. You are not and minimize environmental impact via energy con- optimization but also many new alone in wondering, “Exactly what is the smart grid?” At the servation, i.e., by switching to green and renewable technical challenges. figure 1. Scope of the smart grid. (Image courtesy of ABB.) National Governors Association convention in February 2009, energy sources. We can only meet this objective by the CEO of a major utility began his speech with the confes- increasing energy efficiency, reducing peak demand, What Makes sion that he didn’t really know what “smart grid” meant (see and maximizing the use of renewable energy. the Smart Grid Smart? “Why the Smart Grid Industry Can’t Talk the Talk,” in the The growing consensus in the industry and among many Before we attempt to answer this Traditional Grid Smart Grid “For Further Reading” section). It’s little wonder that people national governments is that smart grid technology is the question, let’s look at a familiar sometimes confuse the smart grid with smart meters and answer to these challenges. This trend is evidenced by but much simpler example and advanced metering infrastructure (AMI) or with interoperabil- the specific provision and appropriation of multi-billion- identify the basic components of ity in communications. dollar amounts on the part of the U.S. government (in the an engineering system that we In this article we offer our perspective on the smart grid. 2009 stimulus program, for example) for research and generally consider to be “smart.” We will look at the drivers for the smart grid, sketch out its development, demonstration, and deployment of smart Consider the cruise-control func- scope, discuss what makes the smart grid smart, and envi- grid technologies and the associated standards. The Euro- tion found in most automobiles. sion its distinguishing features. We will focus on the techni- pean Union and China have also announced huge levels You just set a desired speed and cal challenges that a smart grid must deal with and postulate of funding for smart grid technology research, demonstra- leave the control of the gas pedal to the capabilities a smart grid must have in order to meet those tion, and deployment. the cruise-control function. Once challenges successfully. Due to limitations of space, we will The objective of transforming the current power grid set, cruise control does its job not delve into the details of how to meet such challenges, but into a smart grid is to provide reliable, high-quality electric autonomously for you. It keeps the we will sketch a potential example in order to give technically power to digital societies in an environmentally friendly and engine power output steady when Traditional Grid Smart Grid inclined readers an idea of the technologies that are emerging sustainable way. This objective will be achieved through the the car is on a level road, increases Centralized Generation Generation Everywhere on the horizon. application of a combination of existing and emerging tech- the engine output when the car is Power Flows Downhill Power Flows from Everywhere nologies for energy efficiency, renewable energy integration, going uphill, and reduces engine Utility Controls Connections Anyone May Participate Why Do We Need the Smart Grid? demand response, wide-area monitoring and control, self- output when the car is going down- Behavior: Predictable Behavior: Chaotic An electric grid consists of three main subsystems: the gen- healing, HVDC, flexible ac transmission systems (FACTS), hill. All this is done automatically, eration sources (various power plants); the delivery system and so on. without the intervention of the figure 2. Transition from the present grid to a smart grid.

12 ieee power & energy magazine march/april 2010 march/april 2010 ieee power & energy magazine 13 The smartness of the smart grid lies in the decision intelligence layer, all the computer programs that run in relays, IEDs, substation automation systems, control centers, and enterprise back offices.

The scope of the smart grid extends over all the intercon- driver. Most of us consider cruise control a smart function of nected electric power systems, from centralized bulk gen- cars. Beyond the basic speed control, high-end automobiles eration to distributed generation (DG), from high-voltage feature collision avoidance capability, using adaptive control transmission systems to low-voltage distribution systems, and radar technologies. from utility control centers to end-user home-area networks, Figure 3 is a schematic diagram of a basic cruise control from bulk power markets to demand response service pro- system. Signals for vehicle speed, driver commands (set speed, viders, and from traditional energy resources to distributed increase or reduce speed, and so on), clutch and brake pedal posi- and renewable generation and storage, as shown in Figure 1. tions, and fuel injection throttle positions are fed to an onboard The transition from the present grid to a smart grid and cruise-control computer. There control programs work on the the key differences between the two can be illustrated by input data continuously and, based on control theory algorithms, Figure 2. One can see there is a fundamental shift in the de- sign and operational paradigm of the grid: from central to dis- tributed resources, from predict- able power flow directions to unpredictable directions, from a passive grid to an active grid. In short, the grid will be more dy- namic in its configuration and its operational condition, which will present many opportunities for optimization but also many new technical challenges. figure 1. Scope of the smart grid. (Image courtesy of ABB.) What Makes the Smart Grid Smart? Before we attempt to answer this Traditional Grid Smart Grid question, let’s look at a familiar but much simpler example and identify the basic components of an engineering system that we generally consider to be “smart.” Consider the cruise-control func- tion found in most automobiles. You just set a desired speed and leave the control of the gas pedal to the cruise-control function. Once set, cruise control does its job autonomously for you. It keeps the engine power output steady when Traditional Grid Smart Grid the car is on a level road, increases Centralized Generation Generation Everywhere the engine output when the car is Power Flows Downhill Power Flows from Everywhere going uphill, and reduces engine Utility Controls Connections Anyone May Participate output when the car is going down- Behavior: Predictable Behavior: Chaotic hill. All this is done automatically, without the intervention of the figure 2. Transition from the present grid to a smart grid. march/april 2010 ieee power & energy magazine 13 going and whether it is accelerating or decelerating. The control • Vehicle Speed algorithms (applications) are where the smartness of the system • Steering Wheel Controls lies. The control algorithms make intelligent decisions based • Clutch Pedal on the information provided by the communication infrastruc- • Brake Pedal ture, the available controls, and the desired control objective (maintaining constant speed). But without the actuator system, Cruise Control we can only observe the system passively and helplessly, for the Computer actuator system provides the means of actually making changes (differences) in the physical system. After all, it is a physical car, not a virtual one, that the cruise function controls. Vacuum Valve Control The four basic building blocks identified here can easily be mapped to electric power systems, as shown in Table 1. The focus of the industry effort so far has been mostly Vacuum Actuator on the interoperability of the communication and information model, as suggested by the National Institute of Standards Cable-to-Throttle Valve and Technology (NIST) Smart Grid Interoperability standard Throttle Position road map and the International Electrotechnical Commission Throttle Valve (IEC) documents on smart grid standardization. In “Under- standing the Smart Grid from Definition to Deployment,” the Edison Electric Institute rightly suggested that “advanced figure 3. Schematic diagram of vehicle cruise control. controls provide the ‘smart’ in smart grids.” To borrow a phrase from the real estate business, the three value genera- generate control signals for the vacuum actuator. The vacuum tors for the smart grid are “application, application, and appli- actuator increases or reduces the throttle valve by means of a cation.” To enable smart applications, we need not only good cable. The changes in the throttle valve position change the business logic, control, and optimization theory, we also need engine power output and in turn the speed of the car. new hardware components that can control power flows in the We can identify four essential building blocks necessary network, as well as the output and the consumption of power. in this system: ✔✔ a sensor system to measure system states (automobile What Will the Smart Grid Be Like? speed, brake and pedal position, throttle position) To average consumers, the smart grid, for the most part, will ✔✔ communication infrastructure (wires to collect sensor remain under the hood, working silently and invisibly. Some information and propagate control signals) interfaces will be exposed to consumers, such as the proto- ✔✔ control algorithms (also known as applications that type iPhone interface by which users will be able to check digest the information and generate control signals the current electricity price and electricity consumption and intended to change the state of the system) remotely turn home appliances on and off. Though fascinat- ✔✔ actuators that effect desired changes in the physical ing, such technologies represent only the tip of the iceberg. system (in this example, the throttle valve position and The really important and advanced technologies of the smart the engine power output). grid will remain unnoticed by the general public. All four building blocks are needed for this smart function to When we look beyond the horizon, we envision some work. The sensor system and the communication infrastructure salient features of the smart grid that set it apart from the let the driver know what is going on, i.e., what speed the car is traditional power grid. It is clear that as the system’s supply

table 1. The four building blocks of vehicle cruise control mapped to the electric power system.

Building Blocks Power System Mapping Sensor system Current transformer (CT), voltage transformer (VT), (PMU), , temperature, pressure, acoustic, and so on Communication infrastructure Power line carrier (PLC), wireless radio, advanced metering infrastructure (AMI), home area network (HAN), fiber-optic networks Control algorithms (applications) Wide-area monitoring and control; microgrid management; distribution load balancing and reconfiguration; demand response; optimal power flow (OPF); voltage and var optimization (VVO); fault detection, identification, and recovery (FDIR); automatic generation control (AGC); interarea oscillation damping; system integrity protection scheme (SIPS); and so on Actuator system/physical system HVDC, FACTS, DG, energy storage systems, reclosers, automatic switches, breakers, switchable shunts, on-load tap changers, hybrid transformers, and so on

14 ieee power & energy magazine march/april 2010 Multiple benefits could result from a SIM-based architecture; energy loss would be reduced to a minimum.

and consumption become more decentralized and distrib- Distribution of Production uted, the system’s condition will become more dynamic and DG (from solar, fuel cell, small wind turbine, and other less predictable. The development of demand, supply, and sources) and energy storage (battery, thermal, and hydrogen) power flow control technologies will thus become essential are everywhere in a smart grid. They are not marginal play- in protecting, managing, and optimizing the new grid. In the ers but highly influential and integrated parts of the energy following sections we summarize certain other features of web. They provide energy diversity, reduce demand for cen- the future smart grid. tral fossil-fuel power plants, and increase supply redundancy and system reliability. The distribution of energy production Tightly Integrated Renewable Energy from renewable sources also increases the resilience of the In the smart grid, energy from diverse sources is combined grid in the face of widespread disturbances (e.g., blackouts). to serve customer needs while minimizing the impact on the environment and maximizing sustainability. In addition to A New Level of Controllability nuclear-, coal-, hydroelectric-, oil-, and gas-based genera- In the future smart grid, a new generation of power trans- tion, energy will come from solar, wind, biomass, tidal, and port and control technologies will have become mature other renewable sources. The smart grid will support not only and widely adopted. Current-limiting and current-breaking centralized, large-scale power plants and energy farms but devices based on solid-state technology will help protect residential-scale dispersed distributed energy sources. These valuable grid assets and isolate faults. Power electronics– renewable and green sources will be seamlessly integrated into based transformers will be common. FACTS technology will the main grid. enable system operators to route power flows along the most efficient paths and find the best power production mixes and Proliferation of Energy Storage schedules. Advanced applications in the control center will A smart grid has numerous energy storage centers, large continuously check the state of the grid and determine the and small, stationary and mobile, that it can use to buffer best control strategies from among billions of possibilities the impact of sudden load changes and fluctuations in wind in real time. and solar generation, as well as to shift energy consumption away from peak hours by providing energy balancing, load Real-Time Grid Awareness following, and dynamic compensation of both reactive and Massively deployed sensors will continuously collect end- real power. The recent development of quick-response battery user energy consumption data, weather data, and equipment energy storage systems (BESSs) with voltage source convert- condition and operational status and perform real-time rat- ers (VSCs) has demonstrated the promise and potential ben- ing in the context of actual distribution and transmission line efits of energy storage. flows. The information will be disseminated through highly available, flexible, open (but secure) two-way communica- Growing Mobile Loads and Resources tion infrastructures to any point in the grid where it can be Many loads and resources connected to the future smart grid used to monitor the status of the grid, predict what will hap- will no longer be stationary. Breakthroughs in battery tech- pen next, and develop optimal control strategies. nology are making plug-in electric vehicles (EVs) commer- cially viable. At all times of day, tens of millions of EVs will The Smart Prosumer be connected to the future grid at parking lots near homes, and the Grid-Friendly Appliance workplaces, and shopping malls. These EVs will represent End-user equipment will no longer consist of dumb devices but both mobile loads and potential sources of power. The bat- will form interactive and intelligent nodes on the smart grid. tery systems in these vehicles will be charged or discharged End-user energy management systems will monitor the energy via sophisticated coordination protocols in order to smooth consumption situation in residences, office buildings, and shop- out fluctuations in power demand in different parts of the ping malls. They will know the consumption patterns and pref- grid, avoid power transmission bottlenecks, and render the erences of the occupants, as well as real-time conditions (e.g., grid more stable. Controllers will be able to respond to power market prices, grid stress). They will use the collected infor- system condition signals such as voltages and frequencies as mation to autonomously interact with the grid to determine the well as market signals such as real-time electricity prices. charging and discharging cycles of plug-in electric vehicles, march/april 2010 ieee power & energy magazine 15 schedule washer and dryer cycles, and optimize HVAC opera- to the body’s muscles; the sensor/actuator layer corresponds tions without sacrificing occupants’ comfort. Appliances will to the body’s sensory and motor nerves, which perceive the continuously monitor voltages and frequencies. When the sys- environment and control the muscles; the communication tem experiences distress due to unforeseen disturbances, the layer corresponds to the nerves that transmit perception and appliances will modulate the power consumed to reduce the motor signals; and the decision intelligence layer corresponds stress on the system and help prevent service disruptions. to the human brain. The smartness of the smart grid lies in the decision intel- The Resurgence of DC ligence layer, which is made up of all the computer programs Advancements in materials, power electronics, and sensor that run in relays, intelligent electronic devices (IEDs), sub- technologies will transform the design and operation para- station automation systems, control centers, and enterprise digm of the smart grid. At the generation, transmission, and back offices. These programs process the information col- distribution levels, ac and dc technologies will work together lected from the sensors or disseminated from the communi- harmoniously. HVDC networks embedded in ac networks cation and IT systems; they then provide control directives or will power the world’s megacities but will use only a frac- support business process decisions that manifest themselves tion of the land required for transmission a generation ago. through the physical layer. Some application examples are HVDC transmissions will link clean and renewable power at given below: remote or offshore generation sites to the main power grid. ✔✔ microgrid control and scheduling (demand response Distribution buses in office and residential buildings will sup- and efficiency) ply dc power to digital appliances without the need for power ✔✔ intrusion detection and countermeasures (cybersecurity) adapters. Hybrid grid (ac/dc) architectures for distribution ✔✔ equipment monitoring and diagnostic systems (asset systems will make the grid more flexible and reliable. management) ✔✔ wide-area monitoring, protection, and control Real-Time Distributed Intelligence ✔✔ online system event identification and alarming (safe- In the smart grid, advanced grid-monitoring, optimization, ty and reliability) and control applications track the operating conditions of ✔✔ power oscillation monitoring and damping (stability) grid assets, calculate their ratings, and dynamically balance ✔✔ voltage and var optimization (energy efficiency and load and resources to maximize energy delivery efficiency demand reduction) and security in real time. The increased interactivity among ✔✔ voltage collapse vulnerability detection (security) producers and consumers will mean demand is dynamic ✔✔ autonomous outage detection and restoration (self- rather than static; the grid’s operating environment will healing) appear chaotic, and power flow directions will change in ✔✔ intelligent load balancing and feeder reconfiguration response to market conditions. A new generation of protec- (energy efficiency) tion and control technologies will be called on to maintain ✔✔ self-setting and adaptive relays (protection) the safety and security of both the system and its personnel. ✔✔ end-user energy management systems (consumer par- ticipation, efficiency) The Four Technology ✔✔ dynamic power compensation, using energy storage Layers of the Smart Grid and voltage source inverters (efficiency and stability). The four essential building blocks of the smart grid can be For the decision intelligence layer to work, data (infor- depicted using a layered diagram, as shown in Figure 4. mation) need to be propagated from the devices connected An analogy can be drawn between these layers and those to the grid to the controllers that process the information that make up the human body. The bottom layer is analogous and transmit the control directives back to the devices. The communication and IT layer performs this task. The IT layer serves to provide responsive, secure, and reliable information dissemination to any point in the grid where the information Decision Intelligence is needed by the decision intelligence layer. In most cases, this means that data are transferred from field devices back Communication to the utility control center, which acts as the main repository for all the utility’s data. Device-to-device (e.g., controller- Sensor/Actuator to-controller or IED-to-IED) communication, however, is also common, as some real-time functionality can only be Power Conversion/Transport/Storage/ achieved through interdevice communication. Interoperabil- Consumption ity and security are essential to assure ubiquitous commu- nication between systems of different media and topologies and to support plug-and-play for devices that can be autocon- figure 4. Smart grid technology layers. figured when they are connected to the grid, without human

16 ieee power & energy magazine march/april 2010 intervention. The accelerating deployment of AMI around system architecture can be envisioned, like the one shown the world is a big step in building a two-way communication in Figure 6. The new architecture requires the introduction platform for enabling demand response and other advanced of new building blocks. We shall refer to the new building distribution applications. block technology as the smart integration module (SIM). The physical layer is where the energy is converted, SIMs will have the following functionality: transmitted, stored, and consumed. Solid-state technology, ✔✔ connection to the grid (feeder) power electronics–based building blocks, superconducting ✔✔ ac bus for ac loads materials, new battery technologies, and so on all provide ✔✔ dc bus for dc loads and connection to energy storage fertile ground for innovations. and distributed generation ✔✔ voltage regulation in steady state and in transient Example of a New ✔✔ fast real and reactive power compensation Controllable Component ✔✔ fault detection and fault current limiting and isolation On the journey towards the smart grid, there will be many ✔✔ autonomous distributed intelligent control for short- technology breakthroughs that will have game-changing time-scale control effects on its evolution. What follows is one plausible new- ✔✔ coordination and optimization for longer-time-scale technology scenario, described here as an illustration of one control. of the many potential smart grid technologies that could change the grid’s design and operational philosophy in fun- New Concept of Smart Integration Module damental ways. Advancements in power electronics design and fabrication technology make it possible in principle to design smart inte- Traditional Grid Design gration modules (SIMs) with the aforementioned function- and Premises ality at a competitive cost. If shown to be technologically The design and operation of the traditional power grid is and commercially viable, SIM technology could change the limited by the basic network components (lines, switches, design and operation philosophy and practice in the distribu- transformers with on-load tap changers, switchable capaci- tion and transmission systems in profound ways. tors) available. The traditional grid is built on the following five premises: Benefits and Impacts 1) The components are predominantly dumb conductors Multiple benefits could result from a SIM-based architec- and are not controllable. ture. Energy loss would be reduced to a minimum due to 2) Even if they are controllable, they cannot react quick- ly enough. 3) There is no energy storage; an interruption on the Voltage Regulator transmission or distribution grid means an interrup- tion of service. SS 4) Customer demands are not controllable, and the grid Cap can only react passively to the change in demands ac Bus ac Bus with centralized control. Transformer Transformer 5) The grid can only react to the changes by continuously balancing the output of the central power plants in or- Load Load Load der to remain in a dynamic equilibrium. The lack of energy storage, fast reactive and real power regulation, and distributed generation leads to the traditional figure 5. Traditional distribution system architecture. system design. Figure 5 shows the traditional distribution system architecture. The main function of voltage regulators is to compensate for the voltage drop on the feeder and to maintain feeder voltage SS within acceptable range at the service point. The main function of the switchable capacitor banks is to provide reactive power Load Load Load close to the loads and reduce reactive power flow on the feeder ac Bus and energy losses. SIM dc Bus New Architecture Energy DG Load Load and Enabling Technology Storage With distributed generation, energy storage, and fast- acting converter/inverter technology, a new distribution figure 6. New distribution system architecture with SIM. march/april 2010 ieee power & energy magazine 17 Society in general and the power industry in particular are faced with the challenges and opportunities of transforming the power grid ushered in by Nicola Tesla some 120 years ago into a smart grid.

Conclusion We live in a very critical and exciting time in the evolution of ac Load the electric power industry. Society in general and the power industry in particular are faced with the challenges and UPS ac/dc dc Load opportunities of transforming the power grid ushered in by Battery Nicola Tesla some 120 years ago into a smart grid. A smart grid will help the world manage demand growth, conserve energy, maximize asset utilization, improve grid security figure 7. Traditional solution. and reliability, and reduce its carbon footprint. Smart grid technology is not a single silver bullet but a collection of existing and emerging standards-based, interoperable tech- ac Load nologies working together. Controllable technologies for SIM dc Load supply, demand, power flow, and storage provide the means to implement decisions made by smart control algorithms Battery and thus create value. ABB already provides its customers with many of the smart grid technologies described here and continues to research and develop power control technolo- figure 8. Solution based on SIM technology. gies as well as smart grid applications.

1) maximum utilization of the distributed generation to reduce For Further Reading the real power flow on the grid and 2) provision of reactive J. Berst, “Why the smart grid industry can’t talk the talk,” Smart power where it is consumed to reduce the reactive power flow Grid News, Mar. 5, 2009. on the grid. The power electronics–enabled voltage regula- U.S. House of Representatives, (H.R. 6), Energy indepen- tion capability of SIMs will ensure a high-quality power sup- dence and security act of 2007,” 2007. ply at every load connection point by maintaining optimized US Department of Energy (2008), The smart grid: An intro- voltage levels and compensating for voltage dips, swells, and duction, [Online]. Available: http://www.oe.energy.gov/Smart- flickers. The fast fault current detection and limiting capabil- GridIntroduction.htm. ity will reduce the fault-breaking needs of circuit breakers on European Commission, “European smart grid technology the feeder. The energy storage will provide a short- to medi- platform,” Luxembourg, 2006. um-term power supply buffer so that customer service will Joint US-China Cooperation on Clean Energy (JUCCCE), not be interrupted in the event of short-term disruption on the “Smart grid-future grid?—A basic information report on smart distribution or transmission grid. This will relax the design grid,” Dec. 18, 2007. requirements on the transmission grid. National Institute of Standards and Technology (Sept. 2009), “NIST framework and roadmap for smart grid interop- Data Center Application erability standards,” Release 1.0 (Draft), [Online]. Available: SIM technology could greatly simplify the supplying of http://www.nist.gov/public_affairs/releases/smartgrid_in- power to data centers, a growing market segment, as well as teroperability.pdf improve reliability and reduce energy losses. Figures 7 and Edison Electric Institute, “Understanding smart grid. From 8 illustrate this solution. definition to deployment,” Washington, D.C., Mar. 2009.

Incremental Transition Path Biographies SIM technology is migration-friendly. It can be deployed Enrique Santacana is president and CEO of ABB Inc. incrementally and is compatible with the existing distribu- Gary Rackliffe is vice president of smart grids at ABB Inc. tion system. This is a desirable characteristic, for it allows Le Tang is a vice president and head of the U.S. Corpo- today’s system to be transformed gradually into tomorrow’s rate Research Center at ABB Inc. intelligent system, the smart grid, by changing out tradi- Xiaoming Feng is executive R&D consulting engineer at tional transformers one at a time. ABB Inc.’s U.S. Corporate Research Center. p&e

18 ieee power & energy magazine march/april 2010 Reprinted from March/April 2010 issue of Power & Energy magazine

The Electric Car Is Back (with Help from New Batteries, a Smarter Grid, and Uncle Sam)

By Larry Dickerman and Jessica Harrison

vehicle image ©digital vision, Uncle sam image wikimedia pUblic domain A New Car, a New Grid

Why electrified transportation? Why noW? for decades, inventors, manufacturers, and consumers have tinkered with the idea of electri- fying transportation. to date, however, the global transportation fleet remains fossil fuel based. other technologies such as hydrogen fuel cells and biofuels have also been “on the near horizon” for years. so what makes electrified trans- portation feasible now, and how likely is it that it will transform the transporta- tion sector? let’s look at the facts. ✔✔ investment in electric vehicles (eVs) has moved beyond the fringe, with organizations such as Google spending $10 million for plug-in eV re- search and testing and Warren Buffett investing in a chinese electric car company. W✔✔ the U.s. government has committed to a goal of 1 million plug-in eVs in the next five years and will provide more than $2 billion in stimulus spending for advanced battery development in hybrid electric Digital Object Identifier 10.1109/MPE.2009.935553 systems. march/april 2010 1540-7977/10/$26.00©2010 IEEE ieee power & energy magazine 19 ✔✔ Most major domestic and international original-equip- a battery. The EV has the advantage of mechanical simplicity ment manufacturers (OEMs) are planning to bring but a limited range with present battery technology. The hybrid plug-in EVs to market within the next three years. electric and the plug-in hybrid vehicle have the advantage of Rapid advancement in battery technologies, the warm using the battery to optimize the efficiency of the onboard reception given to hybrid EVs (HEVs) by consumers, and the internal combustion engine. The efficiency improvement is advent of strong policy incentives have all paved the way for accomplished by reducing the size of the internal combus- a transition to electrified transportation, with “how much” tion engine required to achieve expected performance, which and “how quickly” yet to be determined. Potentially large allows tuning around a narrower RPM-torque range. and concentrated new loads, however, mean this develop- Commercial trucks offer some interesting opportunities to ment is worth worrying about. The good news is that the improve efficiency through the use of onboard batteries for concurrent transition of the utility industry to a smart grid powering auxiliary equipment. For example, bucket trucks and the increased use of demand resources can help. used by the utility industry often idle at job sites to provide power for the hydraulic systems that lift and move the bucket. EV Technology Definitions An onboard battery can power the hydraulic system and The term EV has actually come to include several differ- allow the internal combustion engine to be shut off. Figure 1 ent vehicle technologies. The main types available today are shows a hybrid electric bucket truck for utility use. In addi- listed below. tion, other industrial applications of similar technology show ✔✔ HEVs: Hybrid electric vehicles run on gasoline with great promise. For example, any time a load is lifted and let a motor and use batteries to improve fuel efficiency. down with a crane or hoist, using regenerative braking and They do not use electricity from any external source. a battery will enable the system to recapture energy as the ✔✔ Pure-EVs: EVs run on an electric motor powered by load is being let down rather than losing the energy as heat batteries that are recharged by plugging in the vehicle. in braking. ✔✔ Plug-in PHEVs: PHEVs can be charged with electric- ity like EVs and run under engine power like hybrid electric vehicles. The combination offers increased driving range with potential large fuel and cost savings and emission reductions. There are two types: parallel hybrids are PHEVs in which both the electric motor and the combustion engine are mechanically coupled to the wheels through a transmission. Series hybrids, also known as extended range electric vehicles (EREVs), are PHEVs in which the electric motor is directly cou- pled to the wheels and the combustion engine is only used to charge the batteries when needed. In all types of EVs, energy otherwise lost as heat in braking figure 1.✔Hybrid✔electric✔bucket✔truck✔for✔utility✔use.✔ can be recaptured through the use of a generator connected to (Photo✔courtesy✔of✔Dueco.)

table 1. Comparison of transportation technologies.

Pros Cons Gasoline ✔✔ Known✔technology ✔✔ ✔Limited✔availability✔in✔the✔longer✔term✔ Natural✔gas ✔✔ High✔energy✔density✔and✔quick✔refueling ✔✔ Safety✔concerns ✔✔ Domestically✔available✔fuel✔sources ✔✔ Volatile✔price ✔✔ ✔Lack✔of✔roadside✔infrastructure

Diesel ✔✔ More✔efficient✔than✔gasoline✔technology ✔✔ Relies✔on✔foreign✔fuel✔sources ✔✔ Proven✔technology ✔✔ ✔Concerns✔about✔noise✔and✔smell✔ ✔✔ Infrastructure✔already✔exists Fuel✔cell/hydrogen ✔✔ Very✔high✔potential✔efficiency ✔✔ Lack✔of✔hydrogen✔infrastructure✔ ✔✔ No✔local✔emissions ✔✔ Storage✔is✔difficult ✔✔ High✔cost✔and✔short✔equipment✔lifetime PHEV✔or✔EV ✔✔ ✔No✔tailpipe✔emissions✔in✔all-electric✔mode,✔with✔a✔net✔ ✔✔ Battery✔cost✔is✔still✔significant reduction✔in✔CO2✔ ✔✔ Long✔charging✔times ✔✔ High✔efficiency✔and✔performance ✔✔ Some✔additional✔infrastructure✔required ✔✔ ✔Electric✔“fuel”✔is✔widely✔available,✔relatively✔ inexpensive,✔and✔highly✔flexible

20 ieee power & energy magazine march/april 2010 Technology Competition Apart from the economics of the vehicle purchase and As technology has progressed in recent years, operation, EVs and PHEVs have several advantages that may competing technologies have developed as well. A compari- give the technology a strong chance of succeeding: son of the competing technologies is shown in Table 1. ✔✔ The reduction in petroleum usage could significant- The various vehicle technologies available today may not ly reduce dependence on foreign oil, resulting in in- only compete with each other but could actually have strong creased energy security. An HEV with a 40-mi (65-km) synergies. For example, a plug-in hybrid vehicle could have an range could eliminate the need for 2 out of every internal combustion engine powered by a fuel cell or natural gas 3 gal (7.5 out of 11 L) of gasoline consumed by a tra- driving a generator to charge the batteries to extend the range. ditional gasoline vehicle.

✔✔ Such vehicles offer lower net CO2 emissions than cars State of Technology with traditional gasoline engines (electric utility emis- All the necessary technologies to build electric vehicles and sions versus vehicle emissions). plug-in hybrid- electric vehicles exist today. Proof-of-concept ✔✔ They will have no tailpipe emissions (NOx, dust, etc) vehicles and specialty vehicles such as the Tesla sports car when operated in the electric mode. are currently on the road. The battery and control technology ✔✔ EVs and PHEVs can use the existing electric infra- is still being refined, however. Most potential manufactur- structure for charging. ers are looking at lithium-ion batteries as the best balance ✔✔ Such vehicles have the potential to aid the shift to re- between cost, durability, and performance. But the price newable energy sources and make the transport sector premium for lithium-ion batteries will be substantial until more sustainable. volume production brings the price down. ✔✔ Electric vehicle performance is likely to be on a In spite of a higher initial price, electric vehicles and par with or superior to that of similar conventional plug-in HEVs may still look economic to operate due to the vehicles, with better acceleration and very fast re- cost of electricity relative to gasoline. For example, electricity sponse. Their lower center of gravity should enhance at around US$0.10/kWh translates to an equivalent gasoline stability. Lithium-ion technology and other battery cost of about US$0.70 per gallon. With the current US$7,500 advancements can lead to extended ranges in the maximum incentive from the federal government for a pas- electric mode. Fast-charging technology or rapid senger vehicle in the United States, a plug-in hybrid or EV battery change-out could be developed to overcome could still pay back the initial purchase cost within the first many of the issues with range. The vehicles are ex- 25% of its life. With the expected price drop of the batteries tremely quiet. in the coming years, a huge market for EVs may be expected Table 2 shows a partial listing of announced vehicle pro- in the longer term, even without government incentives. duction plans, with a brief vehicle description.

table 2. Sample of vehicle production plans as of November 2009.

Consumer Electric Total Range Battery Manufacturer Model EV Type Availability Range (miles) (miles) Size (kWh) Buick PHEV 2011 10 300 8 Chevrolet Volt EREV Late 2010 40 300 16 Cooper (BMW) Mini E EV Now 156 28 Fisker Karma PHEV 2010 50 300 22 Nissan LEAF EV Late 2010 100 24 Tesla Roadster EV Now 220 53

© GM Corp © GM Corp © GM Corp

march/april 2010 ieee power & energy magazine 21 Potential Electric Utility Impacts is charged when needed. As most vehicles are used for only EVs and PHEVs will increase energy usage on the electric a limited number of hours each day, the benefits offered by utility grid. Given the relatively short duration of the exist- demand response, demand-side management, and even V2G ing peak load on the electric utility infrastructure, EVs and can be very compelling. PHEVs up to very large market penetrations may not require major new investments in generation and transmission if EV Charging customers can be given incentives to charge their vehicles Battery charging will be one of the biggest challenges for during off-peak hours. Initially, however, the impacts on the automotive manufacturers, utilities, customers, and other par- distribution system are likely to be significant. In an indi- ties to work through. The rate of charging will be a matter of vidual home, owning an electric vehicle may mean needing trade-offs. For example, in the United States, a 40-mi-range to upgrade the utility’s local transformer and service drop PHEV might take six hours to charge at 120 V or three hours or perform electrical upgrades in the home. To the extent to charge at 240 V. At 120 V, the utility and the homeowner that multiple EVs or PHEVs show up in a subdivision, other may not require any upgrades to the electrical service. At distribution upgrades may be necessary. 240 V, however, it is almost certain that the homeowner and Coupled with the smart grid and batteries on the distri- perhaps the utility will have to perform service upgrades. bution system, EV and PHEV charging could become part The range of the vehicle will also have a major effect. As of an integrated electric system with the ability to adapt to vehicle range increases, the electrical demand for charging varying conditions. For example, short-term (5 min or less) power will have to increase proportionally to keep charg- interruption of charging could help with electric utility dis- ing times reasonable. Another issue will be the availability tribution system regulation and renewable integration (such of charging stations at work, at shopping locations, or along as residential solar). Interrupting charging for longer periods streets and roads. There are questions about who should of time such as 30 min could also benefit the electric util- provide the charging services (possibilities include utilities, ity by reducing the amount of generation that needs to be commercial establishments, parking garages, employers, kept in spinning reserve. With further development of the and third parties). Then there are questions about how to technology, the batteries in EVs and PHEVs could be used handle the billing for charging away from home. Solutions to temporarily provide power to a home in an outage or even could be as simple as swiping a credit card or as complex as to provide power back to the utility. This feature will not be having the charge added back to the customer’s home utility available with first-generation EVs and PHEVs, however. bill. Tracking usage for road taxes will also be necessary at The batteries used in EVs and PHEVs are an excellent some point. size and technology for small local-storage installations. A Perhaps the most important charging issue is the safety local-storage battery of this type at a home could be charged of the car owner and the general public. Cords running from during off-peak hours, which in turn would allow the vehicle vehicles to plugs in houses or on roadside charging stations to be charged during peak hours via discharge of the local could lead to electrical and tripping hazards. battery. The result would be reduced impact on the utility The electric vehicle supply equipment (EVSE) consists of at peak hours as well as a local battery that can be used for a supply device, a power cord, and a connector. backup in outages. ✔✔ Supply device: This device is the main component In general, one can distinguish three ways to control the of the electric vehicle charging station. Typically it electricity charging process for the electric vehicles: supplies electrical power and provides shock pro- ✔✔ Demand response: influencing the moment of charg- tection; it may also contain information systems for ing, e.g., in response to pricing signals measuring the amount of energy delivered while an ✔✔ Demand-side management: active management of EV is charging. For Level 1 and 2 charging (see defi- the electricity demand, e.g., to prevent overloading of nitions below), the actual charger is located onboard the grid the EV. ✔✔ V2G: the possibility of discharging the battery and ✔✔ Power cord: This is a cable that carries electrical cur- supplying energy to the grid, e.g., at times when elec- rent and communication signals from the supply de- tricity prices are high. vice to the connector. For Level 1 and 2 charging, this These can be very interesting ways for utilities to prevent cord conducts alternating current from the EVSE to problems in the grid, but utilities can also use the storage the onboard charger. capacity available in electric vehicles to incorporate larger ✔✔ Connector: This is a plug on the power cord that con- amounts of renewables. With large numbers of vehicles an nects the EVSE to charging sockets on the electric ve- extended control system will be required, but with such a hicle. In 2010 the Society of Automotive Engineers is system a large amount of grid-connected storage capacity expected to approve SAE J1772, the “SAE Electrical may become available. The user profiles are very suitable Vehicle Conductive Charge Coupler,” as the national for this application, as it is not very important to the end user standard for EVSE connectors, to be used in virtually when the vehicle is charged; it is only important to know it all electric vehicles in the United States.

22 ieee power & energy magazine march/april 2010 Level 1 Vehicle Charging table 3. Typical set of charging options developed for an EV. Level 1 charging is done with a standard outlet and voltage level that is present Charge Power Time to in all homes and businesses. Using this EVSE Utility Service Usage (kW) Charge level of charging may require an upgrade Level 1 110 V, 15 A Opportunity 1.4 18 hours to existing electrical service. Level 1 Level 2 220 V, 15 A Home 3.3 8 hours charging can take 8–14 hours to fully Level 2 220 V, 30 A Home/Public 6.6 4 hours charge an EV, however. For this reason, Level 3 480 V, 167 A Public/Private 50–70 20–50 min Level 1 may not be the customer’s pre- ferred charging method. Level 1 specifi- cations are as follows: be protected from electrical and tripping hazards. A garage, ✔✔ 120-V ac single-phase nominal electric supply carport, or outlet close to a dedicated driveway can likely be ✔✔ 12–16-amp maximum continuous current with made safe by simply installing an EVSE in a wall location 15–20 amps of minimum branch circuit protection. near where the vehicle will be parked. Street charging, how- ever, will require checking to make sure the street right-of- Level 2 Vehicle Charging way is not violated and is used with permission. A special Level 2 charging is faster than Level 1 and is expected to be a charging pedestal will be required, along with appropriate popular option for home charging. Level 2 will often require locking capability, to assure safety. Multifamily homes and an upgrade to existing electrical service and will require a per- apartment complexes will present some of the same chal- manently wired and fixed charging station location. Level 2 lenges as single-family homes with street parking. can fully recharge an electric vehicle in less than half the time required for Level 1. Level 2 specifications are as follows: Charging at Businesses, ✔✔ 240-V ac single-phase nominal electric supply Commercial Sites, and Work ✔✔ 32–70-amp maximum continuous current with 40 amps The concept behind many of the EVs will be to provide a range of minimum branch circuit protection that makes the vehicle useful even if it is only charged at the ✔✔ ground fault protection, no-load make/break inter- place of residence. The availability of charging at the owner’s lock, cable/connector safety breakaway. place of work and at commercial charging sites will however extend the range and value of such vehicles. In some cases, sta- Level 3 Vehicle Charging tions may offer “opportunistic” charging at Level 1 that will be Level 3 charging, also known as fast charging, is a high- as simple as an available 110-V outlet. Much of this charging pow ered technology that can fully charge a vehicle in will probably be offered as a “perk” for employees or shop- 20–30 min. The amount of power required for Level 3 charg- pers. In such cases, the charging may be free to the end user or ing is beyond the capacity of most residential electric service, accounted for by a flat fee that can be paid in a variety of ways. however. For this reason, Level 3 is not expected to be imple- Place-of-work or commercial charging will have the same issues mented for most residential use. In addition, standards for of location and safety as residential charging; the location could Level 3 have not yet been finalized. be at a secured garage or in a parking lot, using a pedestal. Table 3 shows a fairly typical set of charging options developed by an EV manufacturer. Third-Party Charging Stations At least initially, third-party charging stations will be treated In-Home Charging like any other customer served by a utility. A supplier of Table 3 demonstrates why the purchaser of an EV would the service will make app lication for service at a spe- want to pursue charging that would be faster than Level 1. cific location. Any installed facilities will have to meet all Although a purchaser could buy the vehicle and simply plug electrical and construction requirements of the municipal it into a 110-V outlet in a garage, carport, or porch outlet, or county building inspections organization before being the charging time for a nearly discharged vehicle would be served. Since the ideal location for many of the sites will 18 hours. An EVSE with 220-V, 15-A Level 2 charging would be within street rights-of-way, however, a leasing arrangement cut the charging time down to eight hours, and an EVSE with with the municipality may be necessary. In addition, public 220-V, 30-A Level 2 charging would cut the charging time policy around the resale of electricity by third parties in the down to four hours. city will need to be developed. For example, a third-party Level 2 and Level 3 charging will require an approved charging facility may be in an ideal site with little or no EVSE permanently wired at the charging location with an nearby competition. This could put the third-party charg- appropriate home circuit to feed the device. The location ing entity in an unregulated monopoly position that could of the EVSE will vary according to whether the home has be detrimental to end customers. In spite of some potential a garage, a carport, or perhaps just street parking. In all issues, third-party charging stations will ultimately be an cases, the owner of the vehicle and the general public must important part of making EVs successful. march/april 2010 ieee power & energy magazine 23 Commercial trucks offer some interesting opportunities to improve efficiency through the use of onboard batteries for powering auxiliary equipment.

Metering EV Usage ✔✔ Schedule an electrician who can make wiring modifi- Customers will certainly want to take advantage of any cations and install the EVSE. lower rates made available for charging vehicles. Most utili- ✔✔ Understand how to safely charge the vehicle in a man- ties already offer off-peak rates with special meters that ner that avoids electrical and tripping hazards. track the time of use. Many utilities are also considering special off-peak rates for electric vehicles. In the long term, Utility Responsibilities the smart grid will facilitate the tracking of electric vehicle ✔✔ Provide information to the dealer and new owner on usage for billing purposes and possibly for road taxes. Prior how to contact the utility to start the process of prepar- to implementation of the smart grid, another mechanism ing the place of charging to receive the plug-in electric such as a second meter at the customer’s premises, will be vehicle. necessary to track vehicle energy usage at special EV rates. ✔✔ Provide a representative to assist the customer in under- standing the charging options and the needed modifica- Purchasing and Owning an EV tions in the home, business, or other place of charging. The utility industry should recognize that the purchase and ✔✔ Complete any necessary service upgrades after the ownership of an EV will require cooperation and coordi- electrician has completed work and the building in- nation among several stakeholders. The ultimate aim is to spector has issued an approval. make buying and owning an EV as positive an experience as possible. For the new vehicle owner to be satisfied, dealers, Electrician or Electrical the purchaser, the local electric utility, electricians, and city Contractor Responsibilities or county building inspectors will all have important respon- ✔✔ Become certified to install the EVSE and understand sibilities. A few of these responsibilities are listed below. the circuit required to feed the EVSE. ✔✔ Complete the installation of the EVSE and obtain Dealer Responsibilities building inspection approval. ✔✔ Understand and communicate to potential buyers the vehicle specification, performance, and charg- Building and Safety Responsibilities ing options. ✔✔ Provide information to the public on the applicable ✔✔ Clearly communicate to customers the entire process code requirements and the process of obtaining elec- for preparing the home, place of business, and any ad- trical permits, if needed. ditional locations for the charging option the customer ✔✔ Issue electrical permit after all requirements are met. expects to use. ✔✔ Provide information on how to contact each party Federal Incentives involved in the process (utility, electrician, building The American Recovery and Reinvestment Act of 2009 inspection departments). (ARRA) provides energy incentives for both individuals and ✔✔ Provide assistance and tell the customer where to go businesses. The new law modifies the tax credit for qualified for help if problems arise in any part of the ownership plug-in electric drive vehicles purchased after 31 December, experience. 2009. To qualify, vehicles must be newly purchased, have ✔✔ Provide warranty and repair service. four or more wheels, have a gross vehicle weight rating of ✔✔ Provide information on public charging options. less than 14,000 pounds, and draw propulsion using a battery storing at least 4 kWh that can be recharged from an exter- Purchaser Responsibilities nal source of electricity. The minimum amount of the credit ✔✔ Understand vehicle specifications, performance, and for qualified plug-in electric drive vehicles is US$2,500, and charging options. the credit tops out at US$7,500, depending on the battery ✔✔ Determine available electrical outlet options at charg- capacity. The full amount of the credit will be reduced with ing locations (e.g., at home and at the workplace). respect to a manufacturer’s vehicles after the manufacturer ✔✔ Meet with a utility representative to review existing has sold at least 200,000 vehicles. service into the charging location and discuss any ser- Conversions of conventional vehicles to PHEVs receive vice upgrade requirements. tax credits of up to 10% of the cost of conversion. The new

24 ieee power & energy magazine march/april 2010 law provided a tax credit for plug-in electric drive conversion Benefits of a 20-mi Range PHEV kits. The credit is equal to 10% 1,800 of the cost of converting a vehi- kWh cle to a qualified plug-in electric drive motor vehicle and applies Overall to vehicles placed in service after Gasoline Energy CO2 (2006) CO2 (2030) 17 February 2009. The maximum amount of the credit is US$4,000. Electricity The credit does not apply to con- versions made after 31 December 2011. A taxpayer may claim this Approximately credit even if the taxpayer claimed a 32 Approximately 300 Gallons MMBtu 38% hybrid electric vehicle credit for the Approximately same vehicle in an earlier year. A 47% tax deduction of up to US$100,000 per location is available for qualified figure 2. EPRI summary of the benefits of a 20-mile-range PHEV [courtesy of Electric Power Research Institute (EPRI)]. electric vehicle recharging property used in a trade or business. as much load as possible during off-peak hours or allows for short interruptions of charging. For a few utilities, such as Electric Vehicle Environmental Impacts those in southern California, EV penetration will advance PHEV penetration could significantly reduce CO2 emissions much more rapidly. For them, the urgency is great for devel- from the automotive sector and provide a net CO2 benefit oping the appropriate processes and approaches to integra- even when the associated power sector emissions are taken tion as soon as possible. into account. Figure 2 shows an EPRI summary of the ben- efits from a 20-mile-range PHEV, including CO2 reduction. For Further Reading Of course, in any particular situation the generation mix of J. Dowds, C. Farmer, P. Hines, R. Watts, and S. Blumsack, “A the utility providing the electricity will determine the net review of results from plug-in hybrid electric vehicle impact impact. For example, a utility with a high penetration of studies,” technical report, Univ. of Vermont, Dec. 2009. nuclear and renewable energy would provide a greater net EPRI and NRDC, Environmental Assessment of Plug-

CO2 benefit than a utility with primarily coal-fired genera- In Hybrid Electric Vehicles, vol. 1: Nationwide Greenhouse tion. Air quality in urban areas will benefit the most from Gas Emissions, Palo Alto, CA, July 2007. reductions in NOx and volatile organic compounds emis- EPRI and NRDC, Environmental Assessment of Plug-In sions. Depending on the regional off-peak generation mix, Hybrid Electric Vehicles, vol. 2: United States Air Quality emissions of NOx, SO2, and particulate matter could actually Analysis Based on AEO-2006 Assumptions for 2030, Palo increase if they are not already subject to emission “caps.” Alto, CA, July 2007. A. L. Madian, L. A. Walsh, K. D. Simpkins, and R. S. Conclusions Gordon, “U.S. plug-in hybrid and U.S. light vehicle data Diminishing supplies of oil and environmental concerns book:Hybrid vehicles, battery technology, travel patterns, are motivating policy makers to promote practical alterna- vehicle stock, sales trends, performance trends,” in Proc. tives to the internal combustion engine. Advances in bat- Plug-In Electric Vehicles, June 2008. tery technology have already put practical electric vehicles U.S. Government Accountability Office, “Federal energy within reach. Further advancements in manufacturing costs and fleet management: Plug-in vehicles offer potential ben- and storage capacity are needed, however, to make such efits, but high costs and limited information could hinder vehicles appealing to the mass market. The increasing use of integration into the Federal fleet,” June 2009. batteries for utility applications will certainly accelerate the F. Nemry, G. Leduc, and A. Muñoz, “Plug-in hybrid required advancements in technology. In turn, the used bat- and battery-electric vehicles: State of the research and de- teries from vehicles (which can be classified in terms of how velopment and comparative analysis of energy and cost ef- much of the original ability to hold a charge they retain) can ficiency,” Office for Official Publications of the European also be used for residential backup service or for small local Communities, Luxembourg, Rep. JRC 54699, 2009. storage by utilities, thereby helping to reduce the vehicle’s total cost of ownership. In any case, the future of batteries in Biographies vehicles and in electric utilities will have a strong synergy. Larry Dickerman is vice president of T&D Smart Grid In- All utilities will need processes for serving electric vehicles tegration at KEMA Inc. and integrating them into the system in a manner that moves Jessica Harrison is a senior consultant at KEMA Inc. p&e march/april 2010 ieee power & energy magazine 25 Reprinted from May/June 2010 issue of Power & Energy magazine Taking Demand Response to the Next Level

© stockbyte Digital Object Identifier 10.1109/MPE.2010.936352

26 ieee power & energy magazine 1540-7977/10/$26.00©2010 IEEE may/june 2010 To look aT The poTenTial for a smarTer model for The elecTric grid, we should first look briefly at how our current system has been structured. in its feb- ruary 2006 report to congress, the U.s. department of energy (doe) defined demand response (dr) as “changes in electric usage by end-use customers from their normal con- sumption patterns in response to changes in the price of electricity over time, or to incentive payments designed to induce lower electricity use at times of high wholesale market prices or when system reliability is jeopardized.” The model most of us think of—and which has existed for nearly half a century in some form or other—is utility direct load control (dlc). With dlc, the utility sees a system peak and starts cycling off controlled customer equip- ment. This has yielded positive results for utilities, reducing the need in some cases for additional generation. Tas an example, Baltimore Gas and electric (BGe) has been aggressively involved in dr since the 1980s. BGe initiated the rider 5 program in april 1988 for residential and small commercial customers. The program offered customers Us$10 per month during the sum- mer (June–september) for allowing BGe to install a switch on the air conditioner. during particularly hot Leveraging the Experience days, the company would send a signal through a one- way Vhf signal to the switch, and the compressor of U.S. Utilities to cycled on and off such that it operated a maximum of 30 min at any one time. Two years later, the company Demonstrate the Smart offered the rider 6 program, which was a similar type of control for residential water heaters. after nearly Grid’s DR Potential 20 years—of which only a few were spent aggres- sively recruiting customers—the programs had about 250,000 customers enrolled. another example of a dr program is in the city of new Bern, north carolina, a municipal utility that has more than 23,000 load management switches on the system; 10,500, out of a total of 17,210 residential customers, participate in this program. load management is the most difficult type of demand-side management for a utility, involving direct customer participation to cut peak demand through thousands of pieces of equipment. This is also the greenest form of demand-side management: cutting demand at the customer site to offset the need for fossil fuel generation during peak load periods. new Bern is one of the most active load-managing participants in the north caro- lina eastern municipal power agency, with a load management system that has taken more than 20 years to build and that requires constant maintenance to operate. Jon rynne, director of electric utilities for new Bern, says, “over the years, nearly 30% of the switches have failed or been disconnected. once we get into these homes for installation, we can’t get back in to inspect the systems. This is why we are looking at smart grid.” This smart grid would include digital technologies that allow for two way, real-time communications, information, and decisions along the entire grid. These new capabilities have significant positive impact on demand response programs. The federal Utility regulatory commission filed a“ national assessment of demand response potential” with congress in June 2009, projecting that existing dr programs could offset 4% of U.s. peak demand by the year 2019. The report further stated that if these programs were expanded to cover the entire country and a small number of price- responsive programs were added to the mix, the impact would rise to 9%. This report also

By Katherine Hamilton and Neel Gulhar

may/june 2010 ieee power & energy magazine 27 What does this mean for utilities, and what are they doing right now to change the DR model?

projected that 20% of peak demand could be offset by DR ✔✔ Smart energy pricing: BGE piloted various dynamic programs if dynamic pricing programs were deployed to all pricing schemes in 2008 and 2009, including a peak- electric consumers. time rebate and critical peak pricing. These pilot programs focused primarily on residential customers How Are Utilities and gave them price signals one day in advance of Changing Right Now? high-priced periods. Customers were notified through What does this mean for utilities, and what are they doing e-mail, text messages, automated voice calls, and right now to change the DR model? BGE, as one example, is so-called “energy orbs” (globes that would light up planning to offer more customer choices through smart grid green, yellow, or red) that signal periods of high con- technology by sending price signals to customers through sumption by changing color. Through these pricing smart pricing programs (which require a smart grid system). programs and an extensive study conducted with The These pricing signals should provide incentives to customers Brattle Group, BGE learned it could reduce peak de- to make their own investments in DR-type equipment (such mand by more than 25% by simply sending customers as smart appliances) to increase their bill savings. Customer price signals and letting them make rational decisions. empowerment within the DR scheme is key to implementing In addition, these customers reduced their total sum- a smart grid. mer bills by about $100 each on average. BGE plans to Shortly after 2000, DR, conservation and dynamic pric- implement a default peak-time rebate program for all ing started to bubble up as important strategies. There were customers as a part of its overall smart grid rollout. three main drivers: the state of Maryland, determined to avoid ✔✔ Conservation: BGE offers a number of rebate and an energy crisis like the one that happened in California, in cost buy-down programs for residential and commer- 2008 passed the EmPower initiative, requiring the state and cial customers. These include an HVAC rebate, light- utility to reduce electric demand by 15% and achieve 15% ing programs, appliance rebates, and several other electric conservation by the year 2015; the independent sys- initiatives. tem operator, PJM, streamlined the monetization of DR and For New Bern, North Carolina, the smart grid represents conservation programs, allowing utilities like BGE to bid in a way to manage its DR program more effectively. If the capacity in the form of DR to produce a compelling revenue city is able to complete its advanced metering infrastruc- stream; and the federal American Recovery and Reinvest- ture (AMI) pilot project, customers will be able to partici- ment Act allocated US$3.4 billion for the smart grid, award- pate with different levels of involvement to promote energy ing millions of dollars in cost-shared funding to utilities such efficiency and peak demand reduction. The city currently as BGE. As a result, BGE developed a long-term strategy, offers coincidental peak pricing and time-of-use (TOU) Vision2020, that led to the creation of a suite of programs rates to industrial and commercial customers that desire to known as the Smart Energy Saver’s Program. Some of the actively reduce their demand during peak load periods and programs that make up this initiative include: receive a lower cost of power for their participation. With the ✔✔ PeakRewards: This DR program is the next genera- installation of a two-way metering system capable of pro- tion of Rider 5. BGE offers credits to customers for viding real-time load and energy consumption data, these installing a smart switch or smart thermostat. The types of rates can be extended to the residential customers switch or thermostat uses compressor run time to op- as well. Through the development of a residential TOU rate, timize cycling to give BGE a bigger DR impact. Cus- those customers who desire to actively curtail energy con- tomers choosing a thermostat can take advantage of sumption within their homes during peak periods will be a full-featured Honeywell programmable thermostat rewarded through lower energy costs. For customers who do capable of adjusting temperature settings remotely not desire to actively manage usage, the city’s current load through a Web portal. The program began in 2008 and management program may be a more viable option. The has recruited 55,000 new customers while migrating utility’s ability to communicate directly with thermostats 250,000 legacy Rider 5 customers to the new program. and load management switches through a smart meter will The company has bid about 600 MW of capacity into permit the advancement and expansion of the load manage- the PJM market for the 2011 delivery year. ment program. The increased ability to transmit and receive

28 ieee power & energy magazine may/june 2010 Duke’s Energy Internet and Customized Efficiency Duke Energy is launching a pilot, next-generation DR and Next, customer-focused, beyond-the-meter devices will energy efficiency solution called Energy Internet. Energy In- enable new, compelling products and make energy manage- ternet will improve utility resource utilization and increase ment a simple, easy-to-understand proposition for custom- the ability of Duke’s customers to lower their costs and man- ers. These adaptable energy management systems may be age their energy consumption. This solution is enabled by a used to control energy-consuming devices and will allow for second-generation smart grid infrastructure; it will allow tailor-made energy efficiency and DR solutions that reflect customers to optimize their electricity usage by automating customers’ varying priorities, including bill stability, desire to simple, easy- to-understand business and lifestyle choices use the cleanest energy available, and household and business that can save them money. At the same time, the utility will schedules. These systems will also enable Duke to optimize its be able to determine the optimal mix of traditional, renew- generation and delivery of electricity at a more granular level, able, or energy management resources to meet its customers’ improving reliability while lowering costs. Other non- energy- needs. Duke’s Energy Internet vision includes three core ele- related services may be bundled with the energy management ments: advanced, utility-based equipment; customer-focused, systems, allowing customers to manage their businesses and “beyond-the-meter” devices; and the capture and utilization households more effectively. of customer preferences and energy information. These three Last, the Energy Internet will capture customer preferences elements will combine information, telecommunication, and and other energy-related information in order to optimize util- energy technologies in order to revolutionize how utility ser- ity resources, offer new products and services, and manage vices are provided and how customers manage their usage. customer loads to meet regulatory, safety, and financial targets. At the heart of this vision is advanced, utility-based equip- For example, energy prices, weather data, and other energy- ment that will provide Duke with a seamless platform for related information will signal when resources are environmen- using, storing, generating, buying, and selling clean electrons. tally and/or financially costly. Likewise, customer preferences These technologies will integrate Duke’s smart meter infra- about how energy is used at individual homes and businesses structure, data architecture, and customer systems with a will also be factored into how Energy Internet intelligently man- Web-based portal to optimally dispatch utility resources and ages resources. provide operational data that is more robust. Energy Inter- To deploy the Energy Internet pilot program, Duke is planning net also includes the addition of distributed generation and on rolling out the new technology in three stages: the laboratory storage technologies to improve reliability and reduce main- testing of equipment, beta field testing using “friends and family,” tenance costs. and finally a limited customer deployment.

­information­will­let­the­utility­verify­that­the­thermostats­and­ SMUD­only­needs­its­maximum­500­MW­of­capacity­for­ switches­are­operational­and­responding­to­the­load­signal­ 50­hours­each­year,­DR—enhanced­by­smart­grid­technol- sent­by­the­utility­during­periods­of­peak­demand.­This­func- ogies—will­have­a­significant­impact­on­its­ability­to­meet­ tionality­will­greatly­enhance­the­effectiveness­of­the­load­ demand­while­managing­the­costs­to­consumers.­With­the­ reduction­accomplished­by­the­load­management­program,­ implementation­of­distribution­automation,­AMI,­home­area­ allowing­customers­who­do­not­want­to­actively­control­their­ networks­(HANs),­controllable­appliances­and­thermostats,­ usage­during­peak­demand­periods­to­participate­anyway­in­ dynamic­ rates,­ energy­ storage,­ distributed­ and­ renewable­ a­program­that­provides­a­lower­rate­and­reduces­the­need­for­ generation,­and­so­on,­the­stage­will­be­set­for­optimization­ fossil-fueled­ generation­ during­ peak­ demand­ periods.­ The­ of­ the­ grid­ and­ the­ ability­ to­ provide­ operators­ with­ con- use­of­new,­smart­thermostats­for­the­load­management­pro- trol­over­the­electrical­system­down­to­the­circuit,­feeder,­ gram­will­further­enhance­its­effectiveness­and­reduce­the­ and­ even­ the­ customer­ level.­ Parks­ says,­ “With­ a­ combi- misconceptions­held­by­many­consumers­that­load­manage- nation­of­better­customer­communication­through­our­AMI­ ment­will­negatively­impact­their­quality­of­life.­ system,­HAN,­programmable­communication­thermostats,­ The­Sacramento­(California)­Municipal­Utility­District­ and­controllable­appliances,­we­believe­we­will­be­able­to­ (SMUD),­ on­ the­ other­ side­ of­ the­ country,­ has­ big­ plans­ achieve­ significant­ peak­ demand­ reductions­ using­ volun- in­the­smart­grid­arena­over­the­next­several­years.­SMUD­ tary­ customer­ programs.”­ SMUD­ will­ eventually­ expand­ has­committed­itself­to­developing­smart­grid­infrastructure­ its­ efforts­ to­ the­ commercial­ side.­ “The­ smart­ grid­ will­ that­goes­far­beyond­the­scope­of­its­DOE­grant.­Jim­Parks,­ definitely­enable­additional­demand-responsive­programs,”­ program­manager­for­SMUD’s­energy­efficiency­programs,­ Parks­says. says­ that­ the­ smart­ grid­ will­ have­ a­ significant­ impact­ on­ Another­ utility­ with­ big­ plans­ is­ Duke­ Energy,­ whose­ SMUD’s­ability­to­control­loads­during­peak­periods.­Since­ Energy­Internet­initiative­is­aimed­at­integrating­smart­grid­ may/june 2010 ieee power & energy magazine 29 The smart grid enables customer choice and lets customers react during periods of high energy prices.

technology with an easy-to-use, Web-based customer interface Smart energy pricing programs cannot exist without the (see “Duke’s Energy Internet and Customized Efficiency”). network that is the backbone of a smart grid. The dynamic pricing rates that enable additional DR above and beyond How Do the Smart Grid and merely controlling air-conditioning require that smart Information and Communications meters be in place to measure consumption at least once an Technology Affect DR? hour. Because critical system constraints and weather can- While the current model has been for consumers to receive not be predicted well ahead of time, hourly usage data are their bills once a month with little input into or knowl- needed to bill customers on the dynamic rate after the fact. edge about how they use their energy, consumers have now BGE plans to roll out a default peak-time rebate to all its cus- become more sophisticated. People want to be able to con- tomers by 2012. The day before a critical system event, every trol more of their lives from “apps”—applications on their customer will be notified through personal and broadcasted mobile phones, computers, or other devices. They also have messaging. The hourly consumption data collected by the access to more real-time information—how many texts are smart meters and passed through the AMI system and back being sent by their teens, how much money is in their bank to the utility will be used to compute each customer’s rebate. accounts—and it will only be a matter of time and awareness Customers who are highly engaged and motivated will earn of their benefits before consumers demand the same from greater savings than those who are less interested in sav- their electric utilities. Issues such as data management and ing money. The smart grid enables customer choice and lets privacy will need to be considered carefully in a world where customers react during periods of high energy prices. Today, the data have been limited. The Future of Privacy Forum most of BGE’s customers are on a flat-schedule rate: they (www.futureofprivacy.org) has gathered companies who pay the same rate regardless of the true price in the whole- have resolved these issues in the information and communi- sale market during any given day or hour. cations technologies (ICT) sector to begin considering how To further enable smart energy pricing programs and to transfer the lessons learned to the electric utility sector. empower customers, BGE plans to develop an extensive HAN The smart grid data and information platform that will strategy in the next phase of its smart grid rollout. A HAN overlay the electromechanical grid is critical to the evolution builds a network around the smart meter using a relatively of DR and efficiency programs. BGE’s PeakRewards and low-powered radio signal. In the near future, customers will smart energy pricing programs both require the smart grid be able to purchase smart appliances from companies such to be fully functional. Although the PeakRewards program as GE or Whirlpool and configure them to react to periods of is being rolled out systemwide and an extensive marketing high-priced energy. As an example, a smart refrigerator may campaign is already under way, BGE understands the limita- be able to defer the defrost cycle during a high-priced critical tions of a program like PeakRewards without a robust AMI peak window after receiving a signal from the smart grid. network in place. Specifically, the PeakRewards program In addition, HANs enable in-home displays and Web por- still depends on a one-way radio tower system that sends a tals that give customers real-time consumption data directly signal to the switches and thermostats. BGE cannot know from the meter. The customer-friendly display gives the con- which devices are disconnected, unresponsive, or overrid- sumer real information on usage and cost. These smart appli- den by the customer in real time. Further, as BGE learned ances, in-home displays, and other emerging technologies through its legacy Rider 5 program, after several years cus- are important parts of the smart grid and increase the DR tomers find a way to cheat the technology by disconnecting and conservation that can be achieved. And while modify- the device while continuing to receive bill credits for partici- ing customer behavior is an important part of sending price pation. As a result of these concerns and the degradation of signals, many consumers will also want to use technology to DR over time, BGE plans to migrate the PeakRewards assets automate their responses. onto the smart grid network (AMI) so it can monitor and val- Further down the road, with the emergence of plug-in idate customer by customer. This will let BGE maintain high hybrid electric vehicles (PHEVs), distributed generation, system impacts during peak periods for several years and and high-load gadgets and appliances, the utility will need will also offer customers new functionality, such as control- the smart grid to manage its challenges. The proliferation of ling other appliances with the implementation of the HAN. PHEVs could increase the demand on a system dramatically

30 ieee power & energy magazine may/june 2010 if many people come home and plug in at the same time. secretary of commerce, Gary Locke, stated several times Through a smart grid, the utility could put downward pres- at the COP15 negotiations that the smart grid would be the sure on this demand by sending price signals that encourage key enabling platform for ensuring that clean technologies customers to plug in their vehicles during periods when rela- were integrated into the energy ecosystem. Thomas Fried- tively low-cost energy is available. Conversely, a smart grid man recently stated in the New York Times, “Only a market could automatically prevent charging during high-energy-cost shaped by regulations and incentives to stimulate massive periods if the customer so chooses through smart charging innovation in clean, emission-free power sources can make systems. Last, with distributed generation resources like solar a dent in global warming. And no market can do that bet- and wind energy, the smart grid would enable the balancing, ter than America’s.” He went on to assert that clean-energy monitoring, and optimization of the system so customers and technologies, including the smart grid, should be available to utilities could make the best use of these clean sources. all people and countries of all socioeconomic and develop- ment levels. It is up to the United States, however, to use its What Policies and Regulatory tremendous expertise in building and operating its electric Processes Are Needed? grid to take the lead in developing and deploying smarter U.S. policy makers on both the federal and state sides are grid technologies. grappling with how to deal with the rapidly evolving utility industry. On the federal side, Congress is looking at climate What’s Coming Down the Road? and energy legislation that will set national caps on car- Just as in the 1980s most of us could not imagine sending text bon emissions and percentage requirements for low-carbon messages to our colleagues and getting instant responses, energy sources as well as energy efficiency. On the state side, there will be technology innovations on the power grid that we many regulators are still trying to figure out exactly what the cannot foresee. Knowing the history of the telecom industry, smart grid entails and how to fit it into their regulatory pro- we can now envision downloadable energy applications from cesses. On both sides, education is critical to ensure that both appliance manufacturers as well as utilities and aggregators. utilities and their customers can benefit within the regula- Virtual aggregators could be formed by pooling consum- tory structure. Regulatory processes may need to change as ers in neighborhoods and linking their energy use through well. In the 1980s, we saw new and creative rate structures to social media. We are already seeing the “smart home” and incentivize demand reduction in response to enormous load automation as the long-term vision, with air-conditioning growth. Utilities such as Virginia Power explored TOU rates, systems, appliances, sockets, lighting, and so on reacting to thermal storage rates, and standby generation to manage various price signals without any customer intervention. But demand until they could build additional generation capac- only technology coupled with public policy—rate structures, ity to meet booming customer growth. Unfortunately, the low-carbon incentives, and private investment ability—will price of electricity did not create enormous changes on the enable us to move from our traditional, utility-controlled DR customer end. We are now at the “perfect storm,” however, to a system in which DR is one of a host of interactive tech- where prices, the need for low-carbon sources, and contin- nology solutions on a smarter electric grid. ued load growth in the digital age should enable new pricing regimes as well as increased participation in those programs. For Further Reading Time-differentiated pricing, locational and marginal pricing, L. Mansueti, “DOE’s EPACT Report to Congress on Demand and personalized services and rates are all on the table. Response in Electricity Markets,” U.S. Dept. of Energy rep., Mar. 13, 2006. How Will the United States Interview with Jon Rynne, director of Electric Utilities, Position Itself in the Global Market? City of New Bern, NC, about their demand response pro- During the 2009 United Nations Climate Change Confer- gram, Jan. 19, 2010. ence in Copenhagen (COP15), both the International Cham- FERC, “A national assessment of demand response poten- ber of Commerce and the World Energy Council took the tial,” June 2009. position that the environmental mandate of shifting to a Online interview with Jim Parks, SMUD program man- low-carbon future provided a tremendous business oppor- ager, about their smart grid program, Jan. 20, 2010. tunity for technology development. The World Business World Business Council for Sustainable Development, Council for Sustainable Development asserted that “tech- “Towards a low-carbon economy,” 2009. nology is generally not the end goal but a tool to enhance T. L. Friedman, “Off to the Races,” NY Times, Dec. 19, the delivery of revenue and profit-generating activities that 2009. contribute to economic and social development.” It went on to say, “International cooperation has an important role to Biographies play as a catalyst to accelerate technology progress at each Katherine Hamilton is president of the GridWise Alliance. stage.” This technology push certainly includes smart grid Neel Gulhar is a program manager for the Smart Grid technologies, including the next generation of DR. The U.S. Initiative at Baltimore Gas and Electric. p&e may/june 2010 ieee power & energy magazine 31 EducationAdpe.indd 1 1/10/11 4:48 PM Reprinted from May/June 2010 issue of Power & Energy magazine Demanding Standards

Developing Uniformity in Wholesale Demand Response Communications to Enhance Industry Growth

By Scott Coe, Andrew Ott, and Donna Pratt

DemanD response (Dr) is here to stay. What started in a few programs in the wholesale markets just a decade ago has now blossomed into an important compo- nent of the north american wholesale electricity markets. not only are demand resources playing important roles in the forward and real-time markets, these resources are also proving their worth in longer-term capacity markets. to fur- ther the role of demand resources in wholesale electricity markets across north america, some standardization would be desirable. although the standardization process will not address either market designs or technology choices, it does Dhave the goal of developing a common set of data definitions and interactions to facilitate standard communications. Dr success stories are plentiful, and by most accounts there will be many more stories in the future. market opera- tors are relying on demand resources for a larger fraction of the peak demand-fighting “tool kit”—and using these resources more often. in addition to reducing peak load, demand resources are beginning to provide services to the wholesale markets, where they compete with generation supply resources. Demand resource participation will con- tinue to increase with the introduction of new and improved

© digital vision Digital Object Identifier 10.1109/MPE.2010.936350

may/june 2010 1540-7977/10/$26.00©2010 IEEE ieee power & energy magazine 33 technologies in areas such as advanced building HVAC con- While the system operators must go through formal trol systems and local energy storage. stakeholder and tariff filing processes to make changes to As more demand resources have the opportunity and their market designs, they are able to work together to stan- capability to provide enhanced response, resources may dardize various aspects of their operations. The IRC mem- find that meeting the current technological requirements bers believe that they can mitigate some of these differences for participating in wholesale electricity markets creates a through the introduction of improved, coordinated standards high “cost of entry” that can temper this growth. Smaller for demand resources providing energy, capacity, and ancil- resources will have a higher relative cost, and many may not lary services in the wholesale markets. participate. The advent of the smart grid and technological advances in metering and communications, combined with Bridging the Gap: standardization of various aspects of administering DR, will How Standards Can Unify Variations provide opportunities for more demand resources to partici- Developing standards for evolving industries and technolo- pate in wholesale markets. gies is a challenging task. On one hand, standardization pro- According to a study compiled by the ISO/RTO Coun- vides structure to developing markets and the businesses that cil (IRC) in 2009, among the nine largest system opera- transact in those markets. On the other, the standards need to tors in North America there are 48 different opportunities be flexible to incorporate future products and services. for demand resources to participate in wholesale markets, The system operators in North America began their coor- spread among energy, capacity, reserve, and regulation ser- dinated standards activities with an important area for stan- vice. The rules for these opportunities vary greatly, how- dardization: measurement and verification (M&V). M&V is ever, even across the same product types. Examples of such one of the most important areas for DR since it is the basis on differences include the minimum threshold for reduction which demand resources participate in the wholesale mar- capability, the need for real-time telemetry, and the ability kets as supply resources. A demand resource does not have to aggregate resources. Such variations in rules, business generation output; its “response” comes primarily in the processes, and information requirements make it more chal- form of estimating a reduction in the load it would otherwise lenging to develop devices and systems that can be used in have consumed. As part of an effort to develop M&V stan- multiple wholesale markets without significant customiza- dards for DR, the system operators worked together along tion. The reasons for these regional variations may be gath- with other interested members of the industry to recommend ered into three general categories: a standard way to describe DR services, to define DR terms ✔✔ stakeholder input on market designs and event timing (see Figure 1), and to identify five standard ✔✔ varying requirements from regional reliability author- methods of evaluating DR performance (see Table 1). ities (RRAs) These concepts, among others, were developed through ✔✔ reliance on established technical communication an open process facilitated by the North American Energy technologies. Standard Board (NAESB). In March 2009, NAESB members ratified the first ever DR M&V standards for wholesale electricity markets; retail market standards Demand Response Event followed shortly thereafter. As part of the current national Deployment Period focus on the smarter electricity Ramp Sustained Response Recovery grid, the National Institute of Stan- Period Period Period dards and Technology (NIST), as mandated by the Energy In- dependence and Security Act of 2007 (EISA), is coordinating the development of a framework that “includes protocols and model standards for information manage- ment to achieve interoperability of Deployment

Release/Recall smart grid devices and systems.” Normal Operations Reduction Deadline (More information about NIST’s Advance Notification(s) Smart Grid Interoperability Stan- figure 1. NAESB DR event timing. (Source: NAESB. Based on information from dards Project is available at www. “Business Practices for Measurement and Verification of Wholesale Electricity De- nist.gov/smartgrid.) mand Response,” ratified on 2009-03-16. © 2010 North American Energy Standards Since the summer of 2009, the Board. All rights reserved. Reprinted with the express written permission of NAESB.) system operators, in coordination

34 ieee power & energy magazine may/june 2010 with NAESB and NIST, have turned their attention to the table 1. NAESB performance evaluation methods. development of technical standards to address the com­ munications between the market operators and the market ✔ Maximum base load: A performance evaluation participants, including direct participants, such as large methodology based solely on a demand resource’s ability industrial customers and aggregators of retail customers to reduce electricity demand to a specified level, regardless of its electricity consumption or demand at deployment. (ARCs). The standards that will emerge from these initia­ ✔ Meter before/meter after: A performance evaluation tives will have the power to bring standard communications methodology where electricity consumption or demand models to different DR markets, unifying terminology and over a prescribed period of time prior to deployment information requirements. is compared with similar readings during the sustained response period. To begin to narrow the differences, the system opera­ ✔ Baseline type 1: A baseline performance evaluation tors are using a systematic approach to organize and artic­ methodology based on a demand resource’s historical ulate the business processes and data elements the define interval meter data that may also include other variables, DR transactions. such as weather and calendar data. ✔ Baseline type 2: A baseline performance evaluation methodology that uses statistical sampling to estimate the Step 1: Define the Standard Business Practices electricity consumption of an aggregated demand In general, technical standards should be preceded by a solid resource where interval metering is not available for the understanding of current business practices, as it is of great entire population. importance to know the objectives of the standards before ✔ Metering generator output: A performance evaluation methodology, used when a generation asset is located creating them. Even as varied as wholesale electric DR ser­ behind the demand resource’s revenue meter, in which vices are in North America, it is possible to map out a process the demand reduction value is based on the output of the flow that is both specific enough to drive technical standards generation asset. (Based on information from “Business Practices for development and general enough to cover important regional Measurement and Verification of Wholesale Electricity variations. Figure 2 shows the business process model for Demand Response,” ratified on 2009-03-16. © 2010 North wholesale electric DR services. American Energy Standards Board. All rights reserved. Reprinted with the express written permission of NAESB.) While six major subprocesses are identified, the first (market participant registration) and the last (settlements) are not unique to DR services and are not considered to be must reflect the circular nature of collecting data, assessing sys­ a part of the standards currently under development. The tem conditions, and issuing instructions. remaining four processes—shown in the boxes labeled 1 The measurement and performance evaluation process through 4 and described below—will be the focus of the includes the steps necessary to collect demand resource technical standards. meter data and prepare the determinants for settlement. This The enrollment and qualification process includes the process covers many of the concepts documented in the steps required to enroll a DR asset, facility, or resource NAESB M&V standards. in a wholesale market in order to provide one or more DR services. Additionally, some services require that the asset, Step 2: Test the Business Practices facility, or resource demonstrate its capability in advance to With the business processes established, the system opera­ qualify for service. tors created an exhaustive set of “use cases” to test the The forward scheduling and award notification process in ­ cludes the steps from offer sub­ mission to award notification, 1 2 Market Forward Enrollment including the supplemental com­ Start Participant Scheduling and mitment and assessment of reli­ Registration and Award Qualification ability to determine whether Notification demand resources that are enrolled to provide reliability­based (emer­ 4 3 gency) DR should be advised of a Measurement Deployment Settlements End possible future deployment. and and Real-Time Performance The deployment and real-time Communications communication process includes the Evaluation steps for communications to demand figure 2. Summary of the DR business process. (Based on information from “Re- resources providing market­based quirements Specifications for Wholesale Standard DR Signals for NIST PAP09,” voted services on a real­time basis and on by NAESB Smart Grid Task Force on 2010-02-05. © 2010 North American Energy dispatch for reliability­based (emer­ Standards Board. All rights reserved. Reprinted with the express written permission gency) DR. This particular process of NAESB.) may/june 2010 ieee power & energy magazine 35 ­viability­of­the­model.­A­“use­case”­is­a­scenario­in­which­ electric­utilities,­grid­operators,­market­participants,­related­ a­set­of­“actors”­(the­system­operator,­a­market­participant,­ service­providers,­and­consumers.­The­CIM­currently­pro- a­metering­authority,­and­so­on)­interact­with­the­process­to­ vides­significant,­but­not­complete,­coverage­for­informa- determine­whether­the­scenario­can­support­specific­condi- tion­models­related­to­DR­but­may­be­readily­extended­to­ tions.­To­determine­which­scenarios­to­test,­the­system­oper- fill­in­any­functional­gaps.­Indeed,­some­have­already­been­ ators­looked­at­different­permutations­of­the­following­key­ filled­as­a­consequence­of­the­IEC’s­61968-9­standard,­and­ variables,­using­the­NAESB­M&V­terms­and­concepts­as­the­ more­efforts­are­being­planned­to­extend­the­CIM­to­pro- basis­for­the­analysis: vide­the­needed­coverage,­using­the­established­processes­ ✔✔ market products:­energy,­capacity,­reserve,­or­regu- for­CIM­extensions.­ lation To­ perform­ specific­ actions­ or­ provide­ specific­ infor- ✔✔ driver:­economic­or­reliability­ mation­ within­ a­ portion­ of­ a­ business­ process,­ messages­ ✔✔ deployment type:­resource-specific,­bulk,­or­self­(see­ and­corresponding­interfaces­are­often­defined­using­some­ Table­2) selected­subset­of­a­logical­model.­For­DR,­specific­message­ ✔✔ performance evaluation method:­ maximum­ base­ structures­ would­ be­ defined­ using­ selected­ elements­ from­ load,­ meter­ before/meter­ after,­ baseline­ type­ 1­ and­ the­extended­CIM.­The­message­structures­would­then­be­ baseline­type­2,­and­metering­generation­output­(see­ packaged­ using­ appropriate­ integration­ technologies­ with­ Table­1). the­necessary­level­of­security.­One­commonly­used­integra- Each­use­case­can­be­identified­by­combining­a­market­ tion­technology­is­Web­services,­although­other­integration­ product­and­driver,­a­deployment­type,­and­a­performance­ technologies­may­be­needed,­as­in­the­case­of­publication­of­ evaluation­method.­In­this­way,­60­possible­use­cases­were­ notifications­to­large­numbers­of­recipient­devices. generated,­of­which­40­are­feasible­for­DR.­Some­combina- Some­ of­ the­ messages­ being­ defined­ to­ support­ DR­ tions­are­quite­common­in­the­wholesale­markets,­such­as­all­ business­ processes­ include­ the­ following­ information­ variants­of­the­economic­energy­product­utilizing­a­resource- exchanges: specific­deployment­model­and­the­reliability­energy­product­ ✔✔ enrollments­ (requests,­ verifications,­ status­ checks,­ utilizing­ the­ bulk­ deployment­ model.­ Other­ permutations­ approvals,­and­rejections) hint­at­potential­future­market­services;­by­testing­them,­the­ ✔✔ bids­and­offers business­ process­ model­ should­ become­ resilient­ to­ many­ ✔✔ energy­schedules market­evolutions. ✔✔ ancillary­service/capacity­awards One­unique­combination­is­the­economic­energy­product­ ✔✔ forecasts utilizing­the­self-deployment­model,­more­commonly­known­ ✔✔ notifications as­ price-sensitive­ or­ price-responsive­ demand.­ Increas- ✔✔ dispatch­instructions ing­price-sensitive­demand­participation­is­a­major­goal­of­ ✔✔ event­responses wholesale­market­operators,­so­there­is­incentive­to­provide­ ✔✔ baseline­and­meter­data­collection the­relevant­market­information­in­a­standard­format. ✔✔ settlement­determinants. In­the­past,­it­has­been­a­common­practice­to­define­mod- Step 3: Standardize the Information Model els­ and­ interfaces­ on­ a­ project-by-project­ basis,­ as­ only­ a­ The­Common­Information­Model­(CIM)­developed­by­the­ limited­number­of­applications­were­being­integrated­in­an­ International­Electrotechnical­Commission­(IEC)­is­a­logi- otherwise­large­project.­The­need­for­interoperable,­stable,­ cal­information­model­that­defines­a­common­vocabulary­ standardized­interfaces­is­intensifying,­however,­as­DR­inte- for­ the­ integration­ of­ applications­ within­ the­ domain­ of­ gration­extends­beyond­the­enterprise­to­service­providers,­ consumers,­and­devices. table 2. Deployment models. Can Standards Really Work? ✔✔ Resource-specific deployment: The system operator There­is­no­crystal­ball­to­tell­us­whether­or­not­DR­stan- issues dispatch instructions via an established dards­ will­ really­ unify­ the­ industry­ and­ lead­ to­ increased­ communication method to one or more discrete resources designated to provide the DR service. liquidity­of­demand­products­in­wholesale­electricity­mar- ✔✔ Bulk deployment: The system operator issues dispatch kets.­We­can­look­to­similar­initiatives­in­the­past,­however,­ instructions via established communication methods to a and­ if­ the­ factors­ driving­ the­ changes­ are­ similar­ we­ can­ group or block of resources designated to provide the DR expect­similar­results. service. One­of­the­best­examples­of­an­implementation­of­stan- ✔✔ Self-deployment: The deployment of resources is automatic or is initiated by the resource or the DR dards­revolutionizing­the­wholesale­electricity­industry­was­ provider—that is, not by the system operator via a defined the­introduction­of­the­Open­Access­Same-Time­Information­ communication channel. Rather, the resource responds System­(OASIS)­standards­in­1996­with­FERC­Order­889:­ to signals such as real-time electrical system conditions, real-time economic conditions, or market outcomes. “Each­public­utility­…­that­owns,­controls,­or­operates­facili- ties­used­for­the­transmission­of­electric­energy­in­interstate­

36 ieee power & energy magazine may/june 2010 table 3. Drivers for OASIS and DR standards.

Drivers for OASIS Standards Drivers for DR Standards ✔✔ Individual customers wishing to schedule energy on the ✔✔ Individual customers wishing to provide DR services in transmission system had no uniform mechanism for doing so. wholesale markets have no uniform mechanism for doing so. ✔✔ FERC provided guidance through Order 889 to push the ✔✔ NIST is providing guidance through the EISA legislation to industry to action. push the industry to action. ✔✔ System operators were formed to facilitate the ✔✔ System operators are ready for an increase in DR in the transactions. system. ✔✔ Business practices and technical specifications were ✔✔ Business practices and technical specifications are being developed by interested industry members. developed by interested industry members. ✔✔ A standards development organization ratified the ✔✔ There is a plan for standards development organizations to standards and provides a mechanism for improving them ratify the standards and provide a mechanism to improve over time. them over time.

commerce … must develop or participate in an Open Access the short-term productivity losses due to a new system imple- Same-Time Information System.” mentations. Here, as with the euro conversion, some regions Shortly thereafter, any customer wishing to schedule will adopt the standards before others since each risk/benefit energy could make a reservation through one or more OASIS analysis will be unique. Regions with planned upgrades, new reservation systems, known as OASIS nodes. Details for market designs, or unpopular protocols have an opportunity building, operating, and using OASIS nodes were agreed on to be among the first to implement the standards. by members of NAESB and documented in a set of whole- Ultimately, the conversion must take place for DR and sale electric quadrant standards: will require leadership to do so. The benefits of adopting ✔✔ WEQ-001 (Business Practices for OASIS) standards include the growth of industries to support DR ✔✔ WEQ-002 (Business Practices for OASIS Standards and other smart grid initiatives. The system operators can and Communication) only make recommendations on the optimal timing. In many ✔✔ WEQ-003 (OASIS Data Dictionary). cases, guidance from FERC may be necessary to establish a The parallels between what happened with OASIS over quick and efficient time line. As in the case of OASIS adop- a decade ago and what is expected with DR standards are tion, a strong message from FERC and support from the many. Table 3 lists the various drivers for both. industry should make for a winning combination.

The Migration from Individual Acknowledgments Programs to Standard Implementations The authors of this article would like to thank the mem- As planned, the standards for DR communications will be bers of the ISO/RTO Council for funding the project developed through NIST’s smart grid initiative in 2010. team responsible for drafting the standards. The project Converting to these standards will not be an overnight pro- team has representation from all of the system operators, cess, however. with support from KEMA and Utility Integration Solu- Consider for a moment the conversion of multiple curren- tions (UISOL). cies in Europe to the single euro currency. There are a large number of variables that go into the complex decision to adopt For Further Reading the euro in a candidate country, both by that country itself (the (2009, Apr. 28). North American wholesale electricity de- costs of the conversion, the economic benefits of a single cur- mand response program comparison [Online]. Available: rency, political gains, and other concerns) and by the European www.iso-ro.org Economic and Monetary Union (the risk or benefit to the union NAESB. (2009, Feb. 16). Business practices for mea- based on the candidate country’s economic stability, EMU and surement and verification of wholesale electricity demand member political gains, and so on). So in this case-by-case pro- response [Online]. Available: www.naesb.org cess, each country comes on board at its own pace and, argu- International Electrotechnical Commission, IEC 61970 ably, when the time is right for both “ends” of the bargain. Part 61970-301. A similar logic may govern DR standards adoption. Adop- FERC Order 889, Apr. 24, 1996, p. x. tion should benefit the industry in the long term; the benefits, however, which are likely to be long-term cost savings and Biographies increased liquidity, will come at a price. The price will be Scott Coe is a vice president at Utility Integration Solutions. the explicit cost of conversion in the form of new software Andrew Ott is senior vice president of markets for PJM. tools, new training, and perhaps even some new hardware Donna Pratt is the demand response market product spe- infrastructure components, as well as implicit costs such as cialist at the NYISO. p&e may/june 2010 ieee power & energy magazine 37 Reprinted from May/June 2010 issue of Power & Energy magazine

© comstock Get Smart

38 ieee power & energy magazine 1540-7977/10/$26.00©2010 IEEE may/june 2010 By T. Joseph Lui, Warwick Stirling, and Henry O. Marcy

EnErgy gEnEration, consumption, and consErvation arE at the root of many of the most pressing issues facing society today. demand contin- ues to rise steadily while the ability to generate and deliver energy is increasing at a much slower rate. in addition, as stated by the secretary of the u.s. department of Energy (doE), steven chu, in his grid Week 2009 presentation, 25% of u.s. power generation and 10% of its distribution assets are associated with electricity generation required during the roughly 400 hours of annual peak-energy-use peri- ods, which represent hundreds of billions of dollars in investments. Furthermore, in the united states, more than half of the electricity produced is wasted due to power generation and distribution inefficiencies, according to 2002 Energy Flow trends data from the doE. very simply, using less energy in our daily lives, making more Eefficient use of the energy we do produce, and reducing global greenhouse-gas emis- sions associated with energy generation are fundamental to our continued collective prosperity and quality of life. it is the thesis of this article that the achievement of energy con- servation and, hence, global emis- Using Demand Response with sion reductions can be significantly accelerated by integrating smart, Appliances to Cut Peak Energy Use, energy-efficient appliances into a “smart” electricity grid—the so- Drive Energy Conservation, Enable called “smart grid.” smart applianc- es shift the paradigm for appliances: Renewable Energy Sources, and appliances are no longer merely passive devices that drive emis- Reduce Greenhouse-Gas Emissions sions but active participants in the electricity infrastructure that can be drawn upon for energy reduction, energy storage, and the optimiza- tion of the electrical grid for greater compatibility with its greenest energy- generation sources. to amplify this latter point: by providing a variable load, smart appliances connected to the smart grid are ideal complements to renewable sources of energy such as wind and solar power, which are inherently variable in supply. We further believe that given proper incentives and control over their smart products, consum- ers will play a key role in reducing peak demand while lowering costs for consumers and businesses and creating more environmentally friendly power generation. a key feature of the smart grid is demand response (dr). as defined by the association of Home appliance manufacturers (aHam) in a december 2009 smart grid white paper, dr refers to a set of scenarios whereby the consumer, utility, or designated third party can reduce energy consumption during peak usage or other critical energy use periods: “the north american Energy stan- dards Board (naEsB) has defined demand response as ‘changes in electric use by demand-side resources from their normal consumption patterns in response to changes in the price of electricity or to incentives designed to induce lower electricity use at times of potential peak load, high cost periods, or when systems reliability is jeopardized.’ ”

Digital Object Identifier 10.1109/MPE.2010.936353 may/june 2010 ieee power & energy magazine 39 What this says is that when it is necessary to reduce peak ✔✔ Pricing must provide incentives to manage energy use demand to avoid the use of high-cost and high-emission more efficiently and enable consumers to save money. power-generating resources or when the utility encounters The way consumers engage with the smart grid is critical. some other issue on the electrical grid that requires the Consumers must be able to choose when and how they want reduction of electricity demand, it can send a signal to the their smart appliances to participate in the smart grid. The home so that the system will reduce its electrical load during offer of financial incentives—through time-of-use pricing or this critical time period. This article describes the develop- other incentive plans—will be the single biggest driver for ment of a system that will reliably and securely accomplish consumers to change their energy consumption habits. The DR as part of the overarching smart grid. beauty of the smart appliances currently being developed is It makes sense to focus on residential electricity use as a that they will empower consumers to obtain direct economic primary means for enabling DR since, according to a 2009 benefits while also providing significant benefits to utilities Electric Power Research Institute (EPRI) study, residential and society at large (i.e., via lower investments and emissions energy use accounts for a full 38% of the total energy con- reductions) without compromising the core performance of sumed in the United States. Home appliances; water heaters; the products. Finally, the success of the smart grid depends and heating, ventilation, and air-conditioning (HVAC) on public-private partnerships and the adoption of an open, systems together represent more than half of this consump- global standard for transmitting and receiving signals from tion, or 21%, of the total energy used in the United States. a home appliance. Developing the smart grid and linking it to smart appliances and other products having DR capabilities will reliably and Smart Grid DR System Architecture predictably reduce appliance electricity consumption in real time. This will create the opportunity for significant increases Overview in energy efficiency and conservation and meaningful reduc- This section describes the smart grid DR system that Whirl- tions in greenhouse-gas emissions. pool Corporation and its partners will develop and demonstrate as part of a smart grid investment grant under the American Requirements for the Smart Grid Recovery and Reinvestment Act. The description will start at AHAM has outlined three primary requirements for the suc- a high level and then discuss in more detail each system com- cess of the smart grid: ponent. We conclude by describing selected use cases. ✔✔ Consumer choice and privacy must be respected; the Figure 1 is a high-level representation of the smart grid consumer is the decision maker. DR system architecture that Whirlpool will demonstrate; ✔✔ Smart grid communications standards must be open, it is referred to as the Whirlpool Smart Device Network flexible, secure, and limited in number. (WSDN). Consumer feedback and control is highlighted at

Computer or Consumer Energy Control In-Home Display

Electronic Internet Domain Communication Open Communication Module Protocol (OCP) WISE Whirlpool- Home Internet Integrated Router Service Smart Device Environment Controller

Smart Meter/AMI

SEP 1.0 / 2.0 Smart Energy Smart Grid Profile Control System

Home Area Network Smart Meter Domain

figure 1. The WSDN architecture.

40 ieee power & energy magazine may/june 2010 It makes sense to focus on residential electricity use as a primary means for enabling DR since residential energy use accounts for a full 38% of the total energy consumed in the United States.

the top of the figure and represents an overarching WSDN The third level of smart grid DR involves knowing the design element. The consumer must always be in control of potential, and coordinating the response from, hundreds to how appliances respond to signals from the smart grid. millions of homes. Ideally, the smart grid DR profile will Figure 1 further depicts three data communications look just like an inverse power generator to a utility or grid domains, including the smart meter domain, the Internet operator. That is, it will be a nearly square wave in nature, domain, and the home area network (HAN). In an April having a predictable amplitude and reliable performance 2009 paper, the Edison Foundation notes that the smart meter over time. Whirlpool Corporation envisions this level of domain represents the tens of millions of networked smart smart grid DR will take place primarily via messaging on meters that are being deployed by utilities as part of building the Internet, as the networks making up the smart meter a so-called “advanced metering infrastructure” (AMI). The domain generally do not have the bandwidth or refresh Internet domain is the public Internet that consumers typi- rates necessary to either cause or coordinate large-scale DR cally access through a variety of broadband service provid- actions in meaningful time frames. For example, in a 2009 ers. The HAN represents the connection of appliances and Silver Springs Network white paper, it is stated that state-of- other smart devices in the home to one another and to both the-art smart meter networks currently being deployed have the Internet and the smart meter domains. In the case of the refresh rates for pricing and energy use information on the architecture being demonstrated by Whirlpool, the HAN order of one data set per hour per meter. will be controlled by a smart device controller (SDC) that Using the WSDN architecture, Whirlpool and its partners hosts applications for monitoring, controlling, and coordi- will be developing and demonstrating these three levels of nating the activities of appliances and other smart devices smart grid DR for a selected set of smart appliances through- on the HAN. It also acts as a central gateway to both the out 2010 and 2011. Internet and the smart meter domains. Putting the Consumer in Control of DR The Three Levels of DR on the Smart Grid The requirement for consumers to control how their appli- It is important to recognize and keep in mind the three levels ances respond to control and pricing signals from the of smart grid DR that must be developed and coordinated on smart grid has been discussed in numerous forums and is a large scale in order to realize benefits from the smart grid. a primary finding in recent consumer research related to At the lowest level is the response of an individual smart the smart grid, such as that undertaken by Litos Strategic appliance to a smart grid control or pricing signal. This Communication and the Continental Automated Building encompasses how an individual water heater, refrigerator, Association. In these reports, it has also been noted that clothes dryer, or dishwasher responds to the smart grid. very few people want to spend time each day understanding The second level of smart grid DR involves understand- and coordinating how their appliances respond to control ing the present usage of and coordinating the responses and pricing signals from the smart grid. Whirlpool’s smart from all the smart appliances and other smart DR products grid demonstration system will use an in-home display and (e.g., solar panels, electric vehicle supply equipment, and so controller to combine real-time energy use feedback with on) in a given home. This is the role of the HAN. While it simple smart grid energy savings response profiles. Many may always be acceptable to a consumer to stop running the have cited real-time energy use feedback as a key element smart water heater for some period of time, it may not be in helping consumers reduce energy use, and this was also acceptable to also adjust the operation of the clothes washer, a key finding in the time-of-use pricing simulation studies clothes dryer, and dishwasher at the same time. The SDC Whirlpool has conducted with consumers. The smart grid operates the HAN and hosts an in-home energy management energy savings profiles are provided as a straightforward system that manages DR coordination for all of the smart means for consumers to control to what degree they par- appliances in the home based on a personal energy savings ticipate in the smart grid, based on their personal priorities, profile defined by the consumer. The SDC connects to the family schedules, and the energy saving incentive options Internet via the consumer’s broadband Internet connection. offered by local utilities or demand aggregators. Essentially, The SDC also connects to the home’s smart meter via the these energy savings profiles define several levels of par- ZigBee Smart Energy Public Application Profile and com- ticipation, from “full participation” to “opt out” and with munications protocol. several gradations in between. These determine the degree may/june 2010 ieee power & energy magazine 41 Consumers must be able to choose when and how they want their smart appliances to participate in the smart grid.

of coordination and the absolute energy savings that will and the translation of this protocol to the Internet and smart be triggered when control and pricing signals are received meter networks via the communications module, or CM); from the smart grid. The objective is to provide consumers the architecture protocol for the SDC; and the architecture with simple smart grid participation options that produce protocol for a variety of Internet-based services, including maximum benefits with little or no compromise in appli- smart grid DR, referred to as the Whirlpool Integrated Ser- ance performance and with no need for regular consumer vices Environment (WISE). Whirlpool is defining an open interaction. The result for consumers will be a reduction in communication protocol (OCP) across the CM, SDC, and their electricity bill with virtually no effort. WISE to provide open standards–based, secure commu- nication links that connect smart appliances to the WISE Smart Device Networking using open and proven IP-based technologies. The OCP will to Enable Smart Grid DR be supported and made available to any smart device manu- As shown in Figure 1, the WSDN consists of three distinct facturer to allow seamless connection of its smart devices to networking domains—the HAN, the Internet, and the smart the WISE. This architecture provides secure, seamless flow meter network. The system is designed to provide seam- and scaling of smart grid DR application messages and data less connectivity and security across each of these network across all levels of the system. The interface to the smart domains. The HAN and Internet are described in terms of meter network, which is not depicted in Figure 2, is expected the ubiquitous TCP/IP architecture protocol layers depicted to be based on the ZigBee Smart Energy Public Application on the far left side of Figure 2. From left to right in the rest Profile, versions 1.0 and 2.0, which many utilities and smart of Figure 2 is the protocol stack for each individual appli- meter manufacturers have adopted. Apart from communica- ance (i.e., both the proprietary protocol within the appliance tions within each individual appliance, the WSDN is built

TCP/IP Architecture WSDN Architecture Protocol Layers Protocol Layers

Appliance CM SDC WISE

OCP Object Model – Common Profile, Plus Custom Profile OCP Stac OCP Stac XMPP XMPP XMPP OCP Stac Application Layer XML SASL XML SASL XML SASL k k k Turn Turn TLS TLS Turn Stunt Stunt Stunt TLS

Host-to-Host TCP TCP TCP Transport Layer

Internet Layer IP IP IP Broadband Appliance-Controlling Protocol Appliance-Controlling Protocol Inter Zigbee Network Wi-Fi Broadband

Interface Wi-Fi net Connection Layer to Internet

figure 2. Protocol architectures for each of the networks in the WSDN.

42 ieee power & energy magazine may/june 2010 on open networking standards, protocols, and applications. Dishwasher Energy Consumption Profile Figure 2 provokes several obser­ vations. Starting at the left with indi­ Main Wash. Heater Is Final Rinse. Heater 1,400 on for Approx. 5 minmin on AAppropprox. 15–20 miminn vidual appliances, the operation and 1,200 HHeatedeated DDrryy.. interface to these appliances will 1,000 Heater Is OOnn continue to be handled with propri­ atts 800 HeaHeaterter Heater On W Heater OOnn etary protocols and algorithms. This 600 On 400 is essentially where each manufac­ 200 turer differentiates the performance 0 and user experience associated with 0:00 0:08 0:16 0:25 0:33 0:41 0:49 0:58 1:06 1:14 1:23 1:31 1:39 1:48 1:56 each appliance. Whirlpool expects Time (hh:mm) the interface from the HAN to each of the individual appliance DR figure 3. Energy use profile during the operation of a typical U.S. residential actions will become standard at the dishwasher. CM, so that appliances from mul­ tiple manufacturers will seamlessly interoperate and perform appliances that regularly consumes significant amounts of as part of the smart grid. For interfacing each appliance to the energy. How these DR algorithms modify the operation of SDC so that communications can be established with both the the machine is critical to achieving consumer acceptance and smart meter and the Internet, Wi­Fi (IEEE Standard 802.11) delivering DR energy savings and societal benefits. Creating is used (for demonstration purposes) as the physical layer of consumer­relevant DR algorithms depends on detailed knowl­ the CM. But CMs may ultimately come in multiple varieties edge of machine performance. Figure 3 shows energy use or with multiple physical layer capabilities embedded in them during a typical operating cycle for a household dishwasher. (e.g., power line carrier, ZigBee, cellular, and so on). Table 1 provides a summary of a number of relevant energy The WSDN SDC will support Wi­Fi, ZigBee, power use statistics, and Figure 4 provides data from a 2008 National line carrier (PLC), and broadband Internet network inter­ face layers. The Wi­Fi will form the HAN with the smart 0.12 appliances; the ZigBee and PLC will connect with the smart meter and the broadband Internet to the consumer’s Internet 0.1 connection. The SDC will be responsible for managing the consumer’s smart grid energy savings profiles, for coordinat­ 0.08 ing the smart grid DR of all the smart appliances and other 0.06 DR products in the home, for communicating the current ercentage price of energy as loaded into the smart meter by the utility, P 0.04 and for communicating via the Internet with the WISE in 0.02 order to provide real­time energy use and DR energy­saving potential information to the utility or demand aggregator. 0 1 4 7 10 13 16 19 22 Individual Appliances as Hour of Day Part of a Smart Grid DR System A key aspect of creating a DR capability is developing figure 4. Time of use and percent of total U.S. power consumer­ relevant DR algorithms for each of the major home consumption profile for U.S. dishwashers (source: NREL).

table 1. Summary of DR opportunities related to shifting peak electricity use during the operation of major home appliances.

Percent of Peak Average Energy Use Shift Total Energy Peak Minimum Power Moving from Load-Shedding Consumed Cycle Energy in Energy in During Max to Min Period Without in Cycle Time Cycle Cycle Cycle Consumption Adverse Consumer Appliance Type (kWh) (hour) (W) (W) (W) (%) Impact (min) Electric clothes dryer 3.0 0.75 6,000 200 3,000 97 20–60 Dishwasher 1.4 1.75 1,180 240 800 80 60–90 Refrigerator 2.1 24 574 20 89 97 40–60

may/june 2010 ieee power & energy magazine 43 dishwasher use will dramatically Dryer Energy Consumption Profile reduce peak energy consumption. 7.0 From a DR algorithm perspec- 6.0 tive, in addition to the opportu- 5.0 nity for complete deferral of the 4.0 Heater Off, 3.0 Heater On Tumble On entire operating cycle, there are

attage (kW) 2.0 also significant power reduction

W 1.0 0.0 opportunities available during the 0:00 0:05 0:10 0:15 0:20 0:25 0:30 0:35 cycle by delaying the final rinse Time (hh:mm) and/or delaying—or perhaps even eliminating—the heated drying figure 5. Energy use profile for a typical U.S. electric clothes dryer. portion of the cycle. Eliminating the heated drying cycle results in a reduction of the absolute amount of energy consumed as well 0.1 as a time-shifted consumption—a double win. Figures 5 and 6, along with Figures 7 and 8 and Table 1, 0.08 provide similar energy and time of use data for residential clothes dryers and refrigerators. As shown in Figures 5 and 0.06 6, electric clothes dryers offer very significant opportunities for shifting peak electricity use, because the difference in 0.04 ercentage

P electricity use between heating the air the clothes are tum- 0.02 bling in and just running the motor to tumble the clothes is well over 5 kW. The typical time of use for clothes dryers in 0 the United States is spread more evenly throughout the day 1 3 5 7 9 11 13 15 17 19 21 23 than is the case for dishwashers, with the peak occurring Hour of Day at about 10 a.m., thereby providing more opportunities for DR participation. From a consumer performance perspec- figure 6. Time of use and percent of total energy used by tive, clothes dryer cycle times and energy consumption are U.S. clothes dryers (source: NREL). nearly linearly related, indicating that some amount of heater on-off cycling upon receiving a DR signal will not dramati- cally affect drying performance but will lengthen the total Renewable Energy Laboratory (NREL) technical report on operating cycle by an amount slightly less than the total time typical usage curves for the U.S. market. the heater is shut off. Periodically turning the dryer heater From these data it can be seen that the residential dish- on and off and thereby lengthening the operating time has washer is an ideal DR appliance because its energy consump- an additional benefit: it reduces total clothes dryer energy tion can be totally deferred by delaying the operation of the consumption by making better use of residual heat. machine to a later time in the evening without causing signifi- With respect to refrigerators, Figures 7 and 8 and Table 1 cant inconvenience to the consumer. Dishwasher usage in the show that there are some opportunities to shift peak electricity United States spikes in the early evening soon after dinner use by moving the defrost and ice-making cycles to off-peak and often coincides with peak electricity demand. Making times. In general, however, the overall DR opportunity for consumers aware of this and incentivizing them to delay their refrigerators is much less than for other household appliances.

Energy Management Possible Energy Consumption Pattern over a 24-h Period and the WISE 700 This section provides an overview Pulse Ice-Making 600 574.7 Compressor on Cycle Defrost of how the network will be used to atts) 500 (Door Openings = Longer Cycle 386.7 400 Cycles) provide energy management ser- 300 274.8 vices to consumers and utilities. In 200 149.4 applying the WSDN to the smart 118.2 attage (W 100 W grid and energy management, there 0 will be a variety of functional ser- 0:00 1:20 2:41 3:56 5:19 6:40 8:01 9:22 10:41 12:0513:21 14:43 16:04 17:25 18:48 20:07 21:28 22:48 vices available. These will include Time (hh:mm) a number of historical energy use and real-time databases; a variety figure 7. Electricity use profile for a typical U.S. refrigerator. of generic customer interaction

44 ieee power & energy magazine may/june 2010 The smart grid DR system described here will deliver significant benefits for consumers, utilities, and society at large.

services, such as subscriber management and authentication, authorization, and accounting (AAA); business application 0.06 modules such as an energy management server (EMS); an interface to third-party applications; and infrastructure (e.g., 0.05 an interface to the utility back-end systems and interfaces 0.04 to the consumer). Some of these will be off-the-shelf, Web- based functional blocks; others will be developed as part of 0.03 ercentage

the smart grid demonstration program. Figure 9 provides a P 0.02 visual depiction of these functional blocks and some of the instances that may be found within them for the smart grid 0.01 energy management application. 0 1 4 7 10 13 16 19 22 Specific Smart Grid Energy Hour of Day Management Use Cases for the WISE With the system functional blocks defined, we can now look figure 8. Time of use and percent of total energy used by at specific examples of using the WISE for performing aspects U.S. refrigerators (source: NREL).

Databases

DBMS for Data-Warehouse DBMS for Transactions Content Historical Historical Billing and Source Device Accounting Subscriber Application Related Data Consumption Device Status Certificates Profiles Interface Data Data Profiles Database

Account Business Application Management Modules AAA

Device Modules ace Subscriber Management Service Management Server Utility Interf ty Network-Level Backend ar Network Energy-Management Interface Billing and Management Server Accounting Server Third-P Server Demand-Response Application tner and and tner r Pa Utility Accounting, Billing, CCP Device Management Server Web Server Clearing Interface

End User Interface Modules figure 9. The functional blocks for the WISE that will be used to provide energy management services for the smart grid. may/june 2010 ieee power & energy magazine 45 By the end of 2011, Whirlpool will be on track to deliver at least 1 million smart appliances to the U.S. market capable of responding to DR signals.

of residential energy management with the smart grid. We 3) The consumer enters an ID and password to log on. will examine how a consumer will interact with the WISE to Data are passed from the smart phone to the subscriber define a smart energy profile and thereby control how the home management server (SMS) through the Web server. responds to signals from the smart grid. We will also discuss 4) The SMS authenticates the user-entered data (ID and how a utility will interact with the smart grid in order to reduce password) against a subscriber profile database. the amount of power required by the grid in real time. 5) If authentication is successful, the SMS retrieves the consumer’s authorization data (i.e., a smart en- Consumer Interaction with the ergy profile). Smart Grid Using a Smart Phone 6) The SMS builds a front-page smart energy profile Consider the actions that occur when a consumer wants to based on the consumer’s authorized services. alter his or her smart energy profile to control how appli- 7) The SMS forwards the front-page content to the ances respond to energy management and pricing signals Web server. from the smart grid. In this case, the consumer will perform 8) The Web server formats the page and forwards it to this task using a Web-enabled cellular telephone or smart the consumer’s smart phone. phone. The process will be as follows: 9) The consumer modifies fields in the smart energy 1) The consumer downloads the WSDN user applica- profile and submits the modification. tion and installs it on a smart phone. 10) The data updates are forwarded to the SMS via the 2) The consumer launches the WSDN management Web server. app on the smart phone. The app automatically con- 11) The SMS validates the received data and modifies nects to the WISE’s Web server (an end-user inter- the consumer’s profile in the subscriber profile data- face module, shown in Figure 9). base accordingly.

Device WISE

Energy Management Energy Management

S-MIME – Encrypts the Actual Data Sent by the Application over Encrypted Connection OCP Messaging OCP Messaging and Media Interface and Media Interface

XMPP XMPP Bus XML SASL SASL – Authenticates the Connection and Authorize Users over the Encrypted Connection SASL t Bus t

Turn/Stunt TLS TLS – Provides a Secured, Encrypted Connection at the Transport Layer Turn/Stunt TLS anspor anspor Tr TCP Tr TCP OCP OCP IP IP

net Internet Wi-FI Zigbee Inter Broadband

figure 10. The three layers of security provided within the WSDN architecture.

46 ieee power & energy magazine may/june 2010 From the perspective of an individual household, time-of-use pricing will enable significant energy bill savings.

12) The SMS builds a notification message targeted to the ✔✔ The end user device statuses are up-to-date in WISE. consumer’s SDC and sends it to the open communica- The process is as follows. Note that each message between tions protocol (OCP) device management server. an app server and the SDC in each home will go through the 13) The OCP device management server keeps a map- OCP device management server. ping table of its registered devices and subscriber 1) The utility detects a heavy load on its grid and sends IDs. It finds the target SDC, encapsulates the noti- a load-shed request to WISE. The request should fication message in the OCP protocol, and forwards specify the number of watts needed and the time the notification message to the SDC. and duration and provide a list of geographical ar- eas (e.g., ZIP codes). Utility-Scale Energy 2) The utility back-end interface module forwards the Management Using the Smart Grid request to the EMS. The following is an example of how a utility may interact 3) The EMS runs through an algorithm, estimates the with the WISE to cause DR actions across hundreds to mil- potential load shed for the requested geographic ar- lions of homes located within its smart grid. There are two eas, and generates a list of energy-curtailment com- preconditions: mands targeted to particular qualified users. The ✔✔ The utility’s back-end interface is up for contracted algorithm is based on several factors including, for utilities to connect. example, the real-time status of all devices in the

table 2. Summary of cyber security risks and mitigation plans.

Scenario Description Impact to (Threat or Vulnerability) System Mitigation Plan Back-End Server Threats The back-end server is compromised High DoS attacks are mitigated by implementing proven approaches, through DoS attacks. such as restricting concurrent connections and the connection rate from clients, and by using encryption key and certificate control in the end-point devices. Internet Domain Threats The network link between the back-end Moderate Since the proposed architecture is based on a distributed- server and user is compromised. computing model, any single point of failure will not affect the whole environment. The intelligence distributed on the SDCs will continue working by means of the energy management schedule and built-in algorithms. When the lost connections are restored, the whole system will be returned to normal. HAN Threats Home appliances are controlled by Low The dual linkage between the consumers and the servers illegitimate sources. (through the Internet domain and the smart meter domain) gives the smart grid control an opportunity to compare energy usage reports received through the Internet domain with data from the smart energy control. Whenever suspicious patterns are detected, the appliance will be isolated and a report will be sent to the consumer. Smart Meter Domain Threats The interface to the utility grid through the Moderate Once again, since the proposed architecture is based on a smart meter network is compromised. distributed-computing model, any single point of failure will not affect the whole environment. Consumer Energy Control Threats The consumer’s interactive device is Low Since the consumer’s interactive device does not connect compromised. directly to the other two control domains (the smart meter domain and the Internet domain), the threat will be localized and the impact to the system will be minimal. may/june 2010 ieee power & energy magazine 47 geographical areas, user consumption preferences, approach to secure each step in the communication and con- and historical data. trol process, from the HAN across the Internet domain and 4) The EMS sends the estimated load shed to the util- smart meter domain to the smart grid control. These open ity. The utility sends back its go-ahead with the esti- security frameworks and protocols encrypt and transport mated load shed. data and messages while protecting connections from tam- 5) The EMS initiates the command distribution pro- pering, theft, and malicious activity. Additionally, this secu- cess by sending out the energy-curtailment com- rity framework allows configuration of various security lev- mands to all targeted SDCs, using multicast. els for different areas of the network, different applications, 6) Each SDC receives the energy-curtailment com- and different types of data on a real-time basis. Since these mand and executes it with the appliances under technologies have already been proven in the public sphere, its management (this is determined by the smart they provide unbreakable security today and the flexibility to energy profile that the consumer has set up). adapt to emerging threats in the future. 7) Each appliance, upon completion of the energy- The security objectives are to provide: curtailment cycle, reports the energy saved back to ✔✔ Confidentiality: to ensure that information is not dis- the SDC. closed unless authorized 8) The SDC sends the command completion message ✔✔ Integrity: to verify that data sent between the appli- back to the EMS. ance and utility cannot be altered or destroyed 9) The EMS, whenever it receives a completion mes- ✔✔ Availability: to ensure that the smart grid system is al- sage from an end-user device, will update the ways available and the system data are safe (the smart subscriber’s database with the energy-saving data, grid system is also protected from denial-of-service, or for verification and accounting purposes. DoS, attacks and viruses that could potentially bring 10) The EMS, after receiving completion messages from the system down or delete files) all targeted devices or after a predefined period of ✔✔ Privacy: to ensure that each participating family or time, summarizes the total energy saved and sends individual maintains control over personal data. this information to the utility. The security design approach has incorporated the fol- lowing elements: Security Using WISE and the Smart Grid ✔✔ Openness: The security protocols and methods are Cyber security is a critical element in the development and constantly tested, analyzed, and improved in the real deployment of a viable smart grid. The proposed architec- world by the wider security community. This security ture employs proven security technology in a multitiered approach can evolve as new threats emerge.

table 3. Expected savings for an individual household, based on the number of loads for which an electric clothes dryer participates in a DR program by shifting use to a time having a lower electricity cost.

Price Reduction of Off-Peak Power

$ 0.09 $ 0.12 $ 0.15 $ 0.18 $ 0.30 $ 0.70

20 $ 4.68 $ 6.24 $ 7.80 $ 9.36 $ 15.60 $ 36.40

40 $ 9.36 $ 12.48 $ 15.60 $ 18.72 $ 31.20 $ 72.80

60 $ 14.04 $ 18.72 $ 23.40 $ 28.08 $ 46.80 $ 109.20

80 $ 18.72 $ 24.96 $ 31.20 $ 37.44 $ 62.40 $ 145.60

100 $ 23.40 $ 31.20 $ 39.00 $ 46.80 $ 78.00 $ 182.00 er Cycles Shifted y 120 $ 28.08 $ 37.44 $ 46.80 $ 56.16 $ 93.60 $ 218.40

140 $ 32.76 $ 43.68 $ 54.60 $ 65.52 $ 109.20 $ 254.80

160 $ 37.44 $ 49.92 $ 62.40 $ 74.88 $ 124.80 $ 291.20 Number of Dr 180 $ 42.12 $ 56.16 $ 70.20 $ 84.24 $ 140.40 $ 327.60

200 $ 46.80 $ 62.40 $ 78.00 $ 93.60 $ 156.00 $ 364.00

220 $ 51.48 $ 68.64 $ 85.80 $ 102.96 $ 171.60 $ 400.40

48 ieee power & energy magazine may/june 2010 ✔✔ Real-world security: The security protocols and Utilities also can expect significant benefits from using methods are in use every day, proving to consumers a smart grid–based DR system that systematically controls and utilities that their systems, appliances, and private energy reduction across millions of homes at a time in a data are secure. coordinated fashion. These benefits include: ✔✔ Modular architecture: Changes to one feature do ✔✔ automatic energy reduction without any inconvenience not affect the rest of the system. For example, updates to consumers to the WISE do not affect security processes (such as ✔✔ precise control of appliance power usage on a network authentication and encryption). Improvements to the level—a powerful facility for sharing the load among security protocols can be implemented without affect- participating consumers ing the functionality or performance of the smart grid. ✔✔ a clear, real-time view of the aggregated demand-side ✔✔ Standards-based architecture: The security archi- energy-saving potential on the network that enables: tecture builds on existing technologies that have been • the prediction of required supply proven in the real world. • the setting of time-of-use and dynamic energy The security architecture is built using the Extensible pricing Messaging and Presence Protocol (XMPP), a framework ✔✔ the ability to minimize the need for purchasing spin- that leverages numerous existing security technologies to ning reserves, thereby lowering costs and significantly lock, encrypt, and authorize each component and link in reducing carbon emissions the system. By applying multiple levels of protection, this ✔✔ the ability to delay the point at which additional gen- solution provides security greater than that used by financial eration capacity must be built. transactions, e-commerce, and other mission-critical tasks All of these are significant new capabilities that will give performed on the Internet today. utilities a level of understanding and control over their oper- Under this security architecture, the security for ations that Whirlpool expects will lead to further efficiencies the communication between the HAN and the SDC is and savings. achieved through three layers: the Transport Layer Secu- Finally, having estimates for the range of time that peak rity (TLS), the Simple Authentication and Security Layer energy use could be delayed without incurring unaccept- (SASL) protocols, and Secure/Multipurpose Internet Mail able consequences for the consumer (shown in Table 1) Extensions (S/MIME). Figure 10 illustrates this multilay- and also the amount of peak energy use that can poten- ered architecture and the steps involved in each of the tially be shifted provides insight into the benefits that three layers. can be obtained on a macro or societal scale. By simply These multilevel security measures cover a wide array of converting the numbers in Table 1 to peak energy savings identifiable and potential security vulnerabilities. Our secu- potential per one million smart appliances of each type, rity solution not only protects assets in the proposed WSDN we can estimate the total peak energy savings potential. architecture but will also help protect the smart grid itself. The results for these calculations and their extrapolation to Table 2 summarizes the different cyber security risks and their associated mitiga- tion plans. table 4. Economic benefits associated with 1 million smart appliances of each type participating in a DR program. Conclusion The smart grid DR system described here Impact of Moving 1 Million Appliances from On Peak to Off Peak will deliver significant benefits for con- Peak Load Equivalents of Capital Cost Savings Appliance Shifted per Million 500 MW of Constructing sumers, utilities, and society at large. From Category the perspective of an individual household, Appliances Coal Plants Coal Plant* time-of-use pricing will enable significant energy bill savings. Table 3 provides an Dishwasher 1,200 MW 2.4 $ 4.20 billion example of the savings a consumer can expect to realize from participating in DR programs with an electric clothes dryer. Refrigerator 500 MW 1.0 $ 1.75 billion The average U.S. consumer completes approximately 300 loads of laundry per Electric 5,500 MW 11.0 $ 19.25 billion year. If a consumer defers 50% of these Dryer loads to times when there are lower elec- tricity costs, the individual can expect to *Source: Capital cost of coal power: $3,500/kW–Synapse Energy Economics: save from US$40 to more than US$200 Coal Power Plan Construction Costs, July 2008 per year, depending on the price of elec- © 2009 Whirlpool Corporation. All rights reserved. UPA Smart Energy Conference 2009 tricity price from the local utility. may/june 2010 ieee power & energy magazine 49 services/news/doe-secretary-chu-smart- table 5. Environmental benefits associated with 1 million smart appliances of each type participating in a DR program. grid-20090921 (2009, Mar.). Measurement & verification

Environmental Impact–Reduction in CO2 Emissions for demand response programs. Association Percent of Peak Annual Equivalent of Edison Illuminating Companies Load Appliance Demand Move to Reduction In Number of Car Research Committee White Paper p. 8. Category 1 2 Off Peak Hours CO2 Emitted Years of Emission [Online]. Available: http://www.naesb. org/pdf4/dsmee_group2_040909w5.pdf Dishwasher 80% 49.5 Mil Lb 4,200 Electric Power Research Institute. (2009, Jan.). Assessment of achievable potential from energy efficiency and demand response Refrigerator 95% 6.8 Mil Lb 560 programs in the US (2010–2030) [Online]. Available: http://mydocs.epri.com/docs/ public/000000000001018363.pdf Electric 80% 52.1Mil Lb 4,300 Association of Home Appliance Man- Dryer ufacturers. (2009, Dec.). Smart grid white 1 paper—The home appliance industry’s Reduction in emissions from off-peak consumption: 209 lb CO2 / MWHr–eGrid 2007 summary, Dec 2008 principles & requirements for achieving 2 Annual emissions of a personal car: 5.46 metric tons CO2 / vehicle / yr: a widely accepted smart grid [Online]. US EPA; Feb 2009 Available: http://www.aham.org/ht/a/ © 2009 Whirlpool Corporation. All rights reserved. UPA Smart Energy Conference 2009 GetDocumentAction/i/44191 S. Uckun, “Integrating renewable en- ergy into the power grid,” in Proc. Sus- tainable Urban Management Workshop, capital cost savings based on reduced need for new power Mountain View, CA: NASA Ames Research Center, Jan. generating capacity are presented in Table 4. Similar stud- 9–10, 2009. ies for residential electric hot water heaters—such as that R. Vaswani and E. Dresselhuys. (2009). Implementing undertaken in 2009 by the Peak Load Association—have the right network for the smart grid: Critical infrastructure indicated that the consumer-acceptable DR potential for determines long-term strategy. Silver Springs Networks water heaters can be at least as high as that for electric [Online]. Available: http://www.silverspringnet.com/pdfs/ clothes dryers. SSN_whitepaper_UtilityProject.pdf Further estimating the greenhouse-gas emissions reduc- Litos Strategic Communication. The smart grid: An tion potential offered by shifting peak electricity use for introduction [Online]. US Department of Energy, p. 20. the same set of appliances provides the picture presented in Available: http://www.oe.energy.gov/DocumentsandMedia/ Table 5. These relatively simple analyses indicate that the DOE_SG_Book_Single_Pages(1).pdf electricity economic savings and greenhouse-gas emission Continental Automated Buildings Association State of reductions that can be obtained by using the smart grid and the Connected Home Market Survey 2008 [Online]. Avail- the new capabilities it offers consumers and utilities for able: http://www.caba.org/Content/Documents/Document. shifting peak energy demand are very significant. ashx?DocId=32664 Whirlpool and its partners are committed to making R. Herndon, “Building America research benchmark the smart grid a reality. By the end of 2011, Whirlpool definition,” Nat. Renewable Energy Lab., Tech. Rep. NREL/ will be on track to deliver at least 1 million smart appli- TP-550-44816, Dec. 2008. ances to the U.S. market capable of responding to DR sig- R. F. Troutfetter. (2009). Market potential for water heat- nals. Whirlpool and other companies also are working to er demand management. Peak Load Management Associa- make smart water heaters and smart thermostats available tion [Online]. Available: http://peaklma.com/documents/ within a similar time frame. As long as the requirements WaterHeaterDemandManagement.pdf for smart grid success put forward by AHAM are adhered to and implemented, society can expect to start reaping large-scale benefits from the smart grid within the next Biographies several years. T. Joseph Lui is the global director of connectivity for Whirlpool. For Further Reading Warwick Stirling is the global director of energy and S. Reedy. (2009, Sept. 21). Grid week: DOE Secretary Chu sustainability for Whrlpool. on fighting consumer smart-grid resistance [Telephony Henry O. Marcy is the vice president of global technol- Online]. Available: http://telephonyonline.com/business_ ogy for Whirlpool. p&e

50 ieee power & energy magazine may/june 2010 Reprinted from July/August 2010 issue of Power & Energy magazine Engineering the Future

A Collaborative Effort to Strengthen the U.S. Power and Energy Workforce

© master series & john foxx

Some of uS are old, Some of uS are young, and Some of uS refuSe to acknowledge the difference. at any age, electric power and energy engineers contribute to the sus- tainability of life on this planet and the future growth of technology and society on all fronts. at a time when the u.S. economy is still struggling to employ more people, the power and energy sector worries about new talent to replace retiring experience. this article introduces readers to the Power and energy engineering Workforce Collaborative (PWC), an initiative on the part of Ieee Power & energy Society (PeS). the PWC was created to strengthen the u.S. power and energy workforce needed for the smart grid of the future and related technologies. much of the material included here comes from the document shown in figure 1. as these workforce issues greatly affect the united States, this work is being closely Scoordinated with Ieee-uSa. By Wanda Reder, Anjan Bose, Alex Flueck, Mark Lauby, Dagmar Niebur, Ann Randazzo, Dennis Ray, Gregory Reed, Peter Sauer, and Frank Wayno

Digital Object Identifier 10.1109/MPE.2010.937125

july/august 2010 1540-7977/10/$26.00©2010 IEEE ieee power & energy magazine 51 The focus of the PWC is on to bring innovations and improve- U.S. university education. This is ments to the system. partly because an informal survey This departure of engineering conducted in 2003 of 40 U.S. and expertise will have to be met by hir- 27 non-U.S. academic institutions ing new engineers and with other revealed that U.S. universities on methods, including the implementa- average replaced four retired power tion of knowledge retention systems. engineering faculty members with The engineering workforce of the three new power engineering faculty future will also have to supplement members while the non-U.S. univer- traditional power system knowl- sities on average replaced four retired edge with new skills in areas such power engineering faculty members as communication and information with more than five new power engi- technologies. It is these new skills neering faculty members. that will be needed for the success- The U.S. Department of Energy ful deployment of advanced tech- (DOE) has recognized the need nologies for the smart grid while for strong support of university re - maintaining the embedded aging search, which of course translates infrastructure. into strong support of the university figure 1. The action plan prepared by the education of power engineers. A PWC (available on the IEEE PES Web site as How Do We Know There are 2006 DOE report to the U.S. Con- in April 2009). Workforce and Education gress said: “Today, the power engi- System Problems? neering education system in the United States is at a critical There are various indications. These include the following: decision point. Without strong support for strategic research ✔✔ Over the next five years, approximately 45% of engi- in power systems engineering and without qualified replace- neers in electric utilities will be eligible for retirement ments for retiring faculty, the strength of our nation’s univer- or could leave engineering for other reasons. This sity-based power engineering programs will wane and along could amount to more than 7,000 power engineers in with them, the foundation for innovation in the power sector just the electric utilities alone. This number may have to meet our energy challenges in the 21st century.” to be multiplied by a factor of two or three to get to the While the focus of the PWC’s effort is on U.S. workforce, total number of power engineers needed to sustain the and education issues and the stakeholders that influence the U.S. economy. Table 1 gives a breakdown of potential career choices and quality of education of the next genera- losses over the next five years for all engineers, along tion of U.S. power and energy engineers (see “PWC’s Focus: with selected skilled trades. Workforce Problems and Solutions”), similar efforts should ✔✔ According to a PWC analysis of a PES Power and En- be effective for other countries. University power and energy ergy Education Committee (PEEC) university survey, engineering education sits at the crossroads where young high about 40% of key power engineering faculty members school students make career decisions with lifelong impact. at U.S. universities will become eligible for retirement The K–12 pipeline has the human resources that the industry over the next five years; it is anticipated that about 27% needs. At the transition point (age 18–22), education is the of them will actually retire. With about 170 engineer- primary mechanism for attracting new talent into positions of ing faculty members working full-time in power engi- design, analysis, manufacturing, operations, and planning. neering education and research, this means that some 50 senior faculty members will be retiring. This figure An Aging Workforce Affects does not include the senior faculty members already Industry and Educators working less than full-time in the area. Finally, even Everyone knows that in the United States, baby boomers more faculty will be needed to teach the increased are beginning to retire and leave the workforce. The power number of power engineering students who will need and energy industry will be hard-hit by this trend. Over the to be trained to meet the future demand. next five years, roughly half of its engineers may retire or ✔✔ The pipeline of students into engineering is not strong leave the field for other reasons. These experienced elec- enough to support the coming need. Surveys show that trical engineers provided the expertise needed to design, 1) most young people do not know much about engi- build, and maintain a safe and reliable electric power sys- neering and do not feel confident enough in their math tem. They did the planning and construction management and science skills to consider such a career and 2) an to expand it to serve a growing population. They guided insufficient number of parents encourage their chil- the development of needed operating and maintenance dren to consider an engineering career, particularly practices. They learned and implemented new technologies parents of girls.

52 ieee power & energy magazine july/august 2010 PWC’s Focus: Workforce Problems and Solutions What are the problems? • Aging U.S. infrastructure (including people) Power and Energy • Growing demand for clean electricity Industries • A need for the next technology leaders • A need for a unified voice and effort on power University Power and Energy Engineering workforce issues Education • A need to fill vacancies at U.S. universities Who can solve them? (See Figure S1.) K–12 Pipeline Government • Universities, industry, and government working together • Administrators, executives, leaders figure s1. Everyone needs to work together.

✔✔ Enrollment by university students in power and en- teaming with secondary and postsecondary educational ergy engineering courses is increasing (perhaps fu- institutions and the workforce system to address the need eled by interest in renewable energy systems and for a qualified, diverse workforce. In its 2007 LTRA, green technologies to mitigate climate change). But NERC revisited the issue and confirmed: “The loss of the number of students interested in electrical engi- industry workers and their years of accumulated expertise neering is declining. A shrinking pool of electrical due to retirements is a serious threat to the bulk power sys- engineering students limits the future supply of new tem reliability, exacerbated by the lack of recruits entering power engineers. the field.” ✔✔ Hiring of new power engineering faculty is beginning Meanwhile, the demand for power workers to plan, to grow after many years of insufficient growth even design, maintain, and operate the bulk power system has to replace retiring faculty. As time has passed, how- continued to increase with the growing need for new infra- ever, many historically strong university power engi- structure investments in electric generation, delivery, and neering programs have ended. end-use technologies and the rising need for technology innovation, driven by a world beset with new challenges The Impact on Reliability and an industry in transition. With the advent of new renew- These “workforce shortage” considerations and their able technologies to meet the challenges of greenhouse-gas impending impact on reliability have been a recurring regulation, an unprecedented change in the resource mix theme in the recent assessments of long-term reliability and complementary technology innovation will be required. (LTRAs) issued by the North American Electric Reliability The need for new infrastructure and technology innovations Corporation (NERC). In its 2006 LTRA, NERC reported presages a steady or possibly rising need for well-trained that, a loss of expertise, exacerbated by the lack of new engineers and workers. recruits entering the field, is one of the more severe chal- Universities, at the heart of the innovation ecosystem, lenges facing reliability today. Further, the need for line must decide where to invest their limited faculty positions. workers, power plant operators, maintenance/repair work- If research and development funding for biotechnology, ers, and pipe fitters and pipe layers has also increased. The information technology, and nanotechnology overshad- Consortium for Energy Workforce Development (CEWD) ows funding for power and energy technology, especially has been addressing these issues with its stakeholders by grid research, then universities will hire faculty in areas

table 1. Retirements and attrition estimates in electric and gas utilities through 2013 (source: Center for Energy Workforce Development).

Potential Attrition and Retirements as a Percentage Estimated Number of Estimated Number of Job Category of Total Workforce Replacements Needed Retirees All engineers 44.7 14,500 10,000 Technicians 49.0 27,000 20,500 Line workers 40.2 29,500 19,000 Nonnuclear plant 47 .6 12,000 9,000 operators Pipe fitters/pipe layers 45.0 8,500 6,500

july/august 2010 ieee power & energy magazine 53 outside of power and energy. A fundamental imbalance that expand the pipeline of students and to build, enhance, in university research funding could jeopardize the reli- and sustain university power engineering programs. ability of our electric grid, not to mention our progress in the areas of energy independence and security, global Maintaining a Strong Workforce competitiveness, environmental stewardship, and quality To ensure that our society has the well-qualified power and of life. It will take a cooperative effort by industry and energy engineers it needs, we believe the following objec- government to meaningfully address this potential reli- tives must be met: ability issue. ✔✔ Develop and disseminate an image of a power engi- The current economic recession will have a serious and neer based on a realistic vision of how engineers will negative impact on the future workforce. As the demand for be solving challenges facing companies, regions, the electricity decreased and access to capital for infrastructure nation, and the world, thereby improving the quality investments tightened, utility companies delayed or canceled of life. Young people want to choose jobs that make a their resource and transmission projects. To cope with short- difference in the world. term financial difficulties, many companies slowed hiring ✔✔ Motivate interest in power and energy engineering of new employees, reduced the workforce, and encour- careers and prepare students for a postsecondary aged older employees to retire early. As a result, when the engineering education in power and energy engineer- economy recovers in the long term, the shortage of qualified ing. Students should be exposed to engineering even employees will become even more critical. before high school. In addition to students, teachers, counselors, and parents must be the targets of career The PWC information. In 2007, IEEE PES, NERC, and the Power System Engi- ✔✔ Make the higher-education experience relevant, stim- neering Research Center cosponsored a National Science ulating, and effective in training high-quality pro- Foundation (NSF) workshop on this subject. The workshop fessional power and energy engineers. Establish and made the following recommendations: maintain a direct link between power engineering ✔✔ Create a single, collaborative voice on solutions to en- and the solution of major challenges facing the United gineering workforce concerns. States and the world. ✔✔ Strengthen the case for extraordinary efforts to build, ✔✔ Increase university research funding to find innova- enhance, and sustain university power engineering tive solutions for pressing challenges and to enhance programs. student education. ✔✔ Envision the future challenges in electric energy sup- The PWC used these objectives to identify goals and ply and demand, and develop an image that will in- stakeholder actions to aid in overcoming the forthcoming crease interest in power engineering careers. challenges facing the engineering workforce. If these actions ✔✔ Stimulate interest in power engineering careers, and pre- are implemented collectively, major progress will be made pare students for a postsecondary engineering education. toward developing the talent and the educational infrastruc- ✔✔ Make the higher education experience relevant, stimu- ture necessary to meet national objectives. lating, and effective in creating high-quality and pro- fessional power engineers. Priority Actions to Meet the Objectives ✔✔ Encourage and support increased university research 1) Double the production of undergraduate and to find innovative solutions and to enhance student graduate students in power engineering. The col- education. laborative estimates that there are about 800 to 1,000 These recommendations were summarized with the fol- undergraduate students graduating in the United States lowing statement: “Sustaining university programs requires each year with interest in electric power engineering recruitment and retention of students and solid research sup- jobs. In addition, U.S. total enrollment in master’s port. The path to achieving sustainable programs is through and doctoral programs in power engineering amounts collaboration among industry, government, and universities to around 550. About 60% of these graduate students and through the establishment of a single, collaborative voice come from outside the United States. While it is dif- regarding how to meet the coming engineering workforce ficult to know exactly what future power engineering challenges and the need for innovative solutions to regional, workforce needs will be, doubling student graduations national, and global energy problems.” over the next five to eight years is the right direction These recommendations were subsequently converted to head in. into the objectives of the PWC. The PES started the 2) Provide US$4 million in funding for undergradu- collaborative to develop industry strategies and solutions ate power engineering scholarships. To attract stu- to the workforce challenge. The collaborative is working dents into the power discipline, US$4 million should for the transformation of relationships among industry, be provided annually to provide funding for 2,000 government, and universities to support ongoing activities undergraduate scholarships to U.S. citizens. These

54 ieee power & energy magazine july/august 2010 scholarships can also be used to attract high school • work to create federal and state policies that facili- graduates into engineering. (Graduate student support tate the initiative. is included in the research funding goal below.) ✔✔ three working groups in the following areas: 3) Create 2,000 internship and cooperative oppor- • outreach and image tunities for students. To en- • education courage students to pursue power and energy careers • research support. and to provide them with real-life experiences, 2,000 internship and cooperative positions should be cre- Outreach and Image Working Group ated. These industry experiences will be an integral The scope of the outreach and image working group has part of an undergraduate and master’s student’s educa- been defined as follows: tional experience. They will help students understand ✔✔ to coordinate with PES in working collaboratively the industry and help industry recruit new engineers. with industry, government, and universities to define 4) Hire 80 new faculty members over the next five the engineering challenges years to replace retiring faculty, to increase ✔✔ to outline the work required of future engineers, wheth- enrollments, and to broaden educational offerings. er they are in industry, government, or universities Currently there are about 170 full-time power engi- ✔✔ to review empirical research on factors involved in neering faculty members in the United States. They choosing an engineering career and assess the status teach courses in electric power systems, electric ma- of those factors in the job market, schools, and so on chines, and power electronics, among other subjects. ✔✔ to anticipate and appeal to generational characteristics If 27% of them retire over the next five years, almost of high school students 50 new faculty members will have to be hired just to ✔✔ to review the work required of future engineers replace them. Senior faculty members are needed to ✔✔ to define an image in cooperation with PES and CEWD mentor new faculty through the tenure process, which ✔✔ to communicate the new image via CEWD’s Commu- typically takes at least five years. With this in mind, nications Council and PES along with the need to expand the number of enrolled ✔✔ to examine research about the people who influence students, 80 new faculty members should be hired student decisions to enter engineering and assess ways over the next five years to place the power engineering to inform and support them education system on track for the 21st century. ✔✔ to coordinate with CEWD and with the U.S. Depart- 5) Raise annual nonequipment funding of university ment of Labor, NSF, national laboratories, and other power engineering research from US$50 million entities to find effective science, technology, engineer- to US$100 million over the next five to eight years. ing, and math (STEM) outreach programs that could Power engineering research is most often reported in be leveraged and supported the following areas: ✔✔ to collaborate with CEWD on regional program • power systems analysis, computation, economics, delivery. and transmission/distribution • power generation and energy development Education Working Group • power electronics. The scope of the education working group has been defined as follows: Organization of the PWC ✔✔ to gather and analyze data on the present and future In order to reach these goals, the PWC has created the fol- workforce, student trends, and the state of the educa- lowing organizational structure: tional infrastructure ✔✔ a management steering committee (six or seven ✔✔ to coordinate efforts with CEWD to obtain industry, people) tasked to government, and university data to make the case for • provide leadership and focus for the PWC increased investment in education • include representation from various sectors ✔✔ to take this case to industry, government, and universi- • include the leaders of the three working groups (see ties, particularly college deans and department heads, below). with the support of the executive council ✔✔ an executive council made up of university adminis- ✔✔ to propose improvements in existing curricula and ini- trators, industry executives, and government leaders tiatives to enhance preparatory skills and interests in (20 to 25 people) tasked to science, math, engineering, and technology • advocate for the PWC’s initiative and exercise lead- ✔✔ to promote information exchange among the CEWD ership in its implementation Education Council, the PEEC, and universities • support and advise the three working groups ✔✔ to identify actions needed for supporting classroom • plan ongoing strategic activities to sustain the instruction, such as development of lesson materials initiative and tools july/august 2010 ieee power & energy magazine 55 ✔✔ to determine ways to improve collaboration among ✔✔ to organize an effort to envision strategic university industry, government, and universities research needs • to encourage university students to enroll and remain ✔✔ to develop strategies to convince key decision mak- in engineering programs ers to support university infrastructure for power and • to support students through scholarships, fellow- energy engineering research ships, job commitment programs, and other assis- ✔✔ to encourage collaboration to achieve innovative solu- tance tions to challenges and increased support of university ✔✔ to work with PEEC to assess and plan advertising of research university programs ✔✔ to work with the executive council to approach gov- ✔✔ to collaborate with CEWD to share best practices ernment and industry in regards to research support with industry about collaboration and college re- needed to build, enhance, and sustain university re- cruiting search programs. ✔✔ to facilitate knowledge creation and transfer using successful educational models and curricula while Proposed Actions for Stakeholders recognizing that the fundamental responsibility for A major barrier in responding to the power and energy engi- education resides with each university neering workforce challenges is ownership of the solutions ✔✔ to promote coordination among universities to those challenges. It is too easy for stakeholders to say that ✔✔ to engage industry, government, and universities in the solution is someone else’s responsibility (e.g., govern- conversations about the desired education and skills ment’s). Some may say that “the market will take care of it,” of future power engineers forgetting that they are part of the market. A partial list of ✔✔ to improve the collaboration that enhances “hands- the owners of solutions is given in Table 2. on” educational experiences. To show that a wide-ranging set of stakeholders can take ownership of solutions, the collaborative has suggested Research Support Working Group actions that could be taken by various interested groups. The scope of the Research Support working group has been Here are some examples of the suggested actions contained defined as follows: in PWC’s 2009 report. ✔✔ to develop the case for university research supported by government and industry, identifying the barriers What Should Universities Do? to and opportunities for increased support Universities need to do the following: ✔✔ to report on funding trends ✔✔ Work toward doubling the supply of power and energy ✔✔ to look at global experiences and trends engineering students. ✔✔ to identify innovation needs ✔✔ Continue enhancing education curricula and teaching ✔✔ to describe the link between education and research techniques to ensure well-qualified job candidates for ✔✔ to take the PWC’s case to industry, government, tomorrow’s energy jobs, as the University of Minne- and universities, particularly college deans and de- sota has done with assistance from the NSF, the Of- partment heads, with the support of the executive fice of Naval Research (ONR), and the Electric Power council Research Institute (EPRI). ✔✔ Increase research in areas that can contribute to meet- ing national objectives. table 2. Owners of solutions to the challenges ✔✔ Get involved in state consortia collaborating to ad- facing the power and energy engineering workforce. dress workforce issues. ✔✔ Conduct seminars and encourage industry informa- • Universities • State regulators tion sessions to interest university students in power • Community colleges • Federal regulators and energy engineering careers. • K–12 teachers, counselors, • National regulatory ✔✔ Build communication and collaboration with industry, administrators. associations • Parents • Congress particularly between industry executives on the one hand • K–12 service organizations • State legislatures and department chairs and college deans on the other. • Research organizations • State workforce agencies ✔✔ Communicate with industry to discover educational • Professional societies • Department of Labor needs that may require innovative approaches (see • Industry associations • Department of Energy “Developing New Skills”). • Employers • National Science • North American Electric Foundation What Should Employers Do? Reliability Corp. • State and national Employers need to do the following: consortia • Center for Energy ✔✔ Maintain workforce development and hiring ac- Workforce Development • And many more tivities in spite of the economic downturn to avoid

56 ieee power & energy magazine july/august 2010 future reliability problems due to massive delayed Actions Taken in the Last Two Years retirements. The PES and the collaborative have taken the following ✔✔ Communicate with university undergraduate students actions between 2008 and 2010. to offer an exciting image of what an energy engineer will be doing that will make a difference and the new Report Published skills that an engineer will be acquiring. The report Preparing the U.S. Foundation for Future Elec- ✔✔ Offer development opportunities to students, such as tric Energy Systems: A Strong Power and Energy Engineer- mentoring, scholarships, internships, cooperatives, ing Workforce, published in April 2009 and shown in Figure senior capstone projects, part-time jobs, and research 1, has been an important accomplishment in and of itself. support (such as sharing data, allowing the testing of Broad-based support for the report’s conclusions was seen innovative ideas, providing access to company engi- in the endorsement by the collaborative’s executive coun- neers for information and guidance, and offering fi- cil, composed of members from utilities, regional transmis- nancial sponsorship). sion organizations (RTOs), manufacturing, academia, the ✔✔ Seek beneficial opportunities through cooperation U.S. Department of Energy, the Federal Energy Regulatory with universities. Talk with faculty members about Commission, and the National Association of Regulatory workforce needs and major business and techni- Utility Commissioners. cal challenges; listen to their education and research plans. Find ways to work together. Coordination with IEEE-USA on ✔✔ Talk with college deans and department chairs Engineering Workforce Issues about strategic corporate and industry challenges, IEEE-USA and PES have been coordinating on workforce the importance of educating students to become policy issues. The collaborative helped IEEE-USA develop a power and energy engineers, and the need for hir- position statement on electricity delivery workforce training ing new faculty. Provide suggestions for maintain- that addressed how US$100 million in American Reinvest- ing a high-quality and cost-efficient educational ment and Recovery Act (stimulus) funds were going to be system. spent on worker training to modernize the grid. Positions ✔✔ Facilitate lifelong learning through innovative in the statement were reflected in the Federal Opportunity programs with community colleges and universi- Announcement. IEEE-USA organized the distribution of ties. Reward participation in professional associa- the PWC’s report to members of Congress and their staff tions. members during a 2009 “fly-in.” ✔✔ Participate in collaborative efforts among industry, government, and educational institutions to address Collaboration with CEWD workforce issues. Find ways to leverage resources to CEWD (www.cewd.org) was formed to help utilities work achieve common workforce development objectives, together to develop solutions to the coming workforce such as building the student pipeline. shortage in the utility industry. Since its major focus has been on skilled technicians, CEWD looked to the PWC to What Should State and assist it in doing more to increase the pipeline of engineer- Federal Utility Regulators Do? ing students. State and federal utility regulators need to do the following: ✔✔ Keep informed on utility workforce issues and regu- Communications with lated utility actions. the Regulatory Community ✔✔ Find and remove any policy barriers or disincentives PWC members have communicated the dimensions of the to utilities’ taking prudent actions to address work- engineering workforce issues to state and federal regu- force needs. lators and to the DOE. One example was in helping to ✔✔ Allow above-the-line rate treatment of prudent work- organize the panel “Preparing the Next Skilled and Inno- force development actions such as providing schol- vative Workforce for the Electric Power Industry” for arships, hiring in advance of retirement, supporting the 2009 National Electricity Delivery Forum sponsored related university research, building community col- by the DOE and the National Association of Regulatory lege programs, creating workforce development and Utility Commissioners. knowledge retention programs, training, and engaging in competitive marketplace hiring practices. PES’s Career Service for the ✔✔ Actively engage in collaborative activities to address United States and Canada electric power industry workforce issues. PES now sponsors a free online career service for power ✔✔ Assess workforce requirements in making regulatory engineering students in the United States and Canada decisions on reliability and major initiatives such as (www.pes-careers.org). Currently, nearly 800 students and building a smart grid. 165 companies have subscribed. july/august 2010 ieee power & energy magazine 57 Video for Promotion to Middle Developing New Skills and High School Students Power engineering curricula are changing. In Figure S2, an Videos provide a good medium for communicating per- electrical engineering masters student at the University of sonal excitement about the power engineering career field. Wisconsin-Madison uses a prototype to test his design of PES has been working with CEWD on a video for young a power electronics system for integrating energy storage students. The video will be available soon on the PES and CEWD Web sites. It will also be available on DVD. devices into a microgrid. New engineering skills are needed for work on the future power grid. Assistance to Popular Online Career Information Resources The PWC is working to improve the power engineering career information at popular Web sites for career infor- mation for K–12 students, such as tryengineering.com and getintoenergy.com. This information includes, in part, identifying universities with power and energy engineer- ing programs.

Plain Talk and Expert Now Education Offerings and Vision Technology changes such as smart grid deployment, national policies that rely on power system enabling technologies, and job losses during the current recession all point to the need for more lifelong learning opportunities in power and figure s2. An electrical engineering master’s student energy engineering. PES’s Plain Talk courses help technical at work. and nontechnical people to better understand power system technologies. Expert Now is a series of engaging and highly interactive online learning courses based on the best IEEE PES-Careers Global, a Work in Progress educational tutorials and workshops from IEEE conferences A prototype of an international PES student career service, around the world. The courses reflect a vision of robust online PES-Careers Global, is being created, with rollout expected educational opportunities for career development. in the coming months. Region 8 is providing leadership in the planning and initial rollout of the service. Future Work of the PWC The PWC is a transitional organization with a focus on engi- Student Job Fairs at neers for power and electric energy careers. The structure Major PES Meetings in the will be modified as needed to reach the PWC’s objectives in United States and Canada the long term. From this point forward, the initiatives and A very successful job fair at the 2008 T&D Conference organizations of others will be most important in achieving and Exposition showed that students and employers value sustainable attention and action on workforce challenges. job fair opportunities. The PES’s education subcommit- The PWC has already increased its emphasis on communi- tee sponsored a similarly successful job fair at the 2009 cation and advisory activities. general meeting. Job fairs are also being held at the 2010 Some of the actions anticipated in the coming year T&D Conference and Exposition and at the general meet- include: ing this summer. ✔✔ exploration of the roles of PES regional leadership and chapters Philanthropic Scholarship ✔✔ continued cooperation with IEEE-USA and the Center and Internship Initiative for Energy Workforce Development In cooperation with the IEEE-USA Foundation, planning ✔✔ discussions on the formation of regional organizations is under way for a philanthropic effort to raise support to support scholarship and internship programs for student scholarships. As envisioned, support will be ✔✔ the rollout of PES-Careers Global sought from individuals, corporations, and foundations, ✔✔ distribution of a video disseminating a new image of among others. Interviews are now being conducted to a power engineer help develop a campaign and program plan. One gift ✔✔ development of a philanthropic giving campaign to fund, the G. Ray Ekenstam Memorial Scholarship Fund, support scholarships has already been established, and the scholarships will be ✔✔ development of a campaign to increase the number of offered in 2011. industry internships.

58 ieee power & energy magazine july/august 2010 Efforts on these last two items are already under way. ✔✔ promoting an image of the engineer as someone who The PWC is seeking to identify potential sources for funding solves problems that make a difference in the world and to identify the mechanisms for administering the funds ✔✔ increasing interest among our young people in math, for scholarships and student internships. science, and engineering A big change is coming in U.S. power and energy engi- ✔✔ improving the quality of high school and postsecond- neering education programs. By the time you read this arti- ary education cle, the winners of the awards from the U.S. Department ✔✔ building, enhancing, and sustaining university power of Energy’s Financial Assistance Funding Opportunity engineering programs, in part through increased re- Announcement (FOA) titled “Workforce Training for the search support. Electric Power Sector” will have been announced. Accord- In assuring future prosperity, there is work to be done ing to the DOE, “the objective of this FOA is to facilitate by industry, government, and educational institutions. To the development of a well-trained, highly skilled electric build from a position of strength, action is required now. The power sector workforce which is vital to implementing a choice is ours to make; the future is ours to lose. national clean-energy smart grid.” Proposals have been submitted from across the industry and education spec- For Further Reading trum. Since the winning proposals are not yet known, the U.S. Power and Energy Engineering Workforce Collab- implications for reaching the PWC’s objectives are not yet orative (2009). Preparing the U.S. foundation for future clear. Future activities resulting from this federal initiative electric energy systems: A strong power and energy en- could include: gineering workforce [Online]. Available: www.ieee.org/ ✔✔ assistance in advertising new programs, new educa- go/pes-collaborative tional modules, and other initiatives Workforce Trends in the Electric Utility Industry. (2006, ✔✔ forums for sharing plans, progress, and lessons learned Aug.). A report to the United States Congress Pursuant to ✔✔ executive sessions on how to sustain successful efforts Section 1101 of the Energy Policy Act of 2005 [Online]. begun with the stimulus funding. Available: http://www.oe.energy.gov/DocumentsandMedia/ Of course, many other opportunities will not be known Workforce_Trends_Report_090706_FINAL.pdf until after the results of the competition are made public. It North American Electric Reliability Corporation. (2007, is exciting that substantial federal resources have been avail- Oct.). 2007 long-term reliability assessment 2007–2016 [On- able to find solutions to these workforce challenges. Our task line]. Available: http://www.nerc.com/files/LTRA2007.pdf is to help make the best solutions sustainable. National Science Foundation, Division of Electrical, One of the most difficult challenges is increasing univer- Communication and Cyber Systems (2007). National Sci- sity research funding, particularly for grid-related research. ence Foundation Workshop on the Future Power Engineer- Enabling technology to support the integration of renew- ing [Online]. Available: http://ecpe.ece.iastate.edu/nsfws/ ables and smart grids along with the expansion of demand Report%20of%20NSF%20Workshop.pdf resource programs while supporting existing and new cen- S. H. Strauss, J. A. Schwarz, and E. Lippmann, “Are tral station generation will require innovative solutions and utility workforces prepared for new demands? Recommen- new technologies. University researchers can make important dations for State Commission inquiries,” Rep. 10-01, Natl. contributions in grid research if funding is available. With Regulatory Res. Inst., Washington, DC, Jan. 2010. funding, universities would also be more likely to hire new faculty to replace retiring faculty. Although the challenges Biographies are great, industry and government funding sources to date Wanda Reder is with S&C Electric. have been insufficient to provide the needed growth in uni- Anjan Bose is with Washington State University. versity research programs on grid-related topics. We hope Alex Flueck is with Illinois Institute of Technology. that through continued discussions with key decision makers, Mark Lauby is with North American Electric Reliabil- ways can be found to solve this problem. ity Corp. Dagmar Niebur is with Drexel University. Conclusions Ann Randazzo is with the Center for Energy Workforce Skilled power and energy engineers are the foundation of the Development. construction of our nation’s future electric energy systems. Dennis Ray is with Power Systems Engineering Re- An ample supply of these engineers is critical to success in search Center. meeting national policy objectives in energy independence Gregory Reed is with the University of Pittsburgh. and security, global competitiveness, environmental stew- Peter Sauer is with the University of Illinois at Urbana/ ardship, and quality of life. To obtain that supply, we have to Champaign. start making investments today by: Frank Wayno is with Cornell University. p&e

july/august 2010 ieee power & energy magazine 59 Continued from page 64

nications and information infrastruc- management of generation and power since the labels get derived from pre- ture that supports a variety of important flow on the grid. These applications are defined objects. applications besides the asset manage- based on data collection from substa- ment functions. Sensing and measure- tions throughout the grid with a scan Widespread Sensors ment technologies are a key enabler of rate ranging from 2 to 4 s. Sensors and actuators in the smart grid the smart grid. They are envisioned as In addition to the EMS system, there are fundamental elements required for being able to support a faster and more is other information available to the successful operation of the grid. It is accurate response to functions such as operator. Some of these are dynamic through these devices that the actual remote monitoring, time-of-use pric- line ratings, synchrophasor data, and power system’s reaction to various in- ing, and demand-side management. a variety of condition puts and outputs is mea- Precise intelligence of the parameters monitoring information. Smart grid- sured. As in classic con- required and their current and forecast The majority of this in- trol systems, it is through values can be used to determine the formation tends to reside enabled asset sensors that vital infor- limitations that various power delivery outside the EMS and may mation about a variety elements impose on any given situation. not be well integrated into management of conditions is received. This added intelligence should provide the overall situational The sensors convert volt- greater clarity and improved decisions. awareness capabilities will be a age, current, phase angle, More detailed examples of functional- in the control center. powerful position status, and other ity and interoperability issues associ- Much of the condition data into manageable ated with supporting this functionality monitoring information tool for signals that are either an- can be found in the NIST Interoperabil- is implemented within alog or digital in nature. ity Framework [1]. standalone systems that electric utility In the power deliv- monitor transformer con- ery smart grid, sensors Smart Grid-Enabled ditions or circuit breaker companies in will increase in both Asset Management conditions. The goal of many ways. type and quantity. In The use of more sophisticated con- the smart grid is to make fact, over the last 10–15 trol algorithms and technologies such monitoring information more generally years, a number of new commercially as expert systems, inference engines, available to a wide range of applications produced sensors have been intro- knowledge bases, and other advanced that would include a broader range of duced by suppliers. These have in- processing approaches have been stud- asset management functions. Data his- cluded transformer monitors, circuit ied for over 20 years. In trying to move torians are examples of data repositories breaker monitors, and infrared camer- forward with these ideas in power de- that provide the infrastructure for this as. Also, EPRI has been very active in livery systems, we have typically run objective. the development and commercializa- into implementation challenges due tion of a whole host of sensors for not to the high cost of communications What Does the Smart only typical substation assets but also systems. There were no cost-effective Grid Offer? for transmission lines. These include mechanisms established to effectively The recent NIST Interoperability sensors like the corona camera, sag- integrate field equipment to enable the Framework provides extensive refer- o-meter, partial discharge, and acous- widespread use of advanced control al- ences to a number of standards that tic sensor. While all of these sensors gorithms. With the smart grid deploy- are suggested to be used as the foun- provide useful information to the util- ment, communications infrastructures dation for the smart grid. Most if not ity operators and managers, they have are being designed based on open stan- all of these standards are designed been difficult to apply out to the field dards that will allow more widespread to have interoperability and self- assets due to the lack of wide area, integration with field equipment for as- description as key elements of their high-bandwidth communications. As set management applications. make up. The interoperability feature a result, many of these remote sen- minimizes the work effort to inter- sors have been limited to periodic Situation Analysis: face applications and data between data collection, local controls that are Where Are We Today? domains. The self-description feature not integrated with the overall grid Today’s transmission systems have further minimizes the labor compo- management and asset management a fair amount of intelligence. EMSs nent by automatically describing a functions, and specialized, proprietary employ advanced applications such as given data element. It also eliminates communication systems. state estimation, contingency analysis, many of the human errors associated One prevalent sensor at every utili- and voltage stability for the continuous with typing and labeling data points ty is the microprocessor relay. Each of

60 ieee power & energy magazine november/december 2010 Information

External Analytics CIM-Based Analytics Applications Data Warehouse

Data Integration Utility Applications Pub/Sub Middleware Data Communication

IP Enabled Digital Communications Network

Grid Data Sources Gateway Gateway Automated Substations LAN Zigbee Wireless Sensor Sensor Net Smart Distributed Equipment Smart Digital Sensor Line Monitoring Sensors Manhole Platform Sensors Meter Network Relays

figure 1. Conceptual integrated sensor data.

these relays contains a significant number tensive (including the installed base Conclusion of data values that go beyond what is of microprocessor based relays). The Smart grid-enabled asset management necessary for protection. Included in limitation to fully utilize this data for will be a powerful tool for electric utili- many of these devices are digital fault both operations and asset management ty companies in many ways. It is one of records, sequence of event recorders, has been limited bandwidth connecting the five fundamental technologies that calculation of I2t, synchrophasor mea- the field assets to the utility enterprise will drive the smart grid, according to surement, and breaker contact status and lack of standards for sharing and the U.S. Department of Energy (along and timing. System-wide communica- managing the data. with integrated communications, sens- tions infrastructure and standard inter- ing and measurement, advanced con- faces for collecting and managing this Data Integration trol methods, and improved interfaces information will make this valuable Integrating the data from these with decision support). data source available to a wide range widely dispersed sensors and intel- Asset management functions are of applications. A simple example: by ligent devices is the key to success made possible through the imple- accumulating the I2t value for each for advanced applications. Figure mentation of a smart grid conceptual circuit breaker over time, an algorithm 1 illustrates the general concept of model and architecture that facilitates can trigger maintenance of the breaker utilizing the smart grid to enable as- interconnection within and between at a prescribed value that would be in- set management functions. Grid data power system domains with standard dicative of contact wearing. Also, by sources are enabled by a widespread interfaces that facilitate interoper- monitoring the contact timing values communication network. Standards ability. Asset management functions one could identify breakers that oper- for data integration are based on the primarily cut across the operations, ated slowly and once again perform IEC Common Information Model transmission, and distribution domains maintenance. Simple targeted main- (CIM) so that the data can be shared of the conceptual model. Asset-man- tenance can prevent costly device by many applications across an en- agement functions will take advantage failures and also minimize costs by terprise service bus. New analytics of widespread sensors that are integrat- maximizing the value of the mainte- will process and evaluate this infor- ed with operational functions in each nance efforts. mation to optimize the performance of the domains but can be used for a The influx of sensors into the elec- and maintenance of assets through- wide variety of functions even across tric utility marketplace has been ex- out the grid. domains through interoperability november/december 2010 ieee power & energy magazine 61 62 ieee power & energy magazine november/december 2010

Standards.indd 1 1/13/11 4:19 PM ­standards­like­the­CIM.­CIM­and­many­ and­ circuit­ breakers.­ These­ research­ For Further Reading other­standards­to­facilitate­interoper- efforts­should­analyze­the­data­sourc- NIST­ framework­ and­ roadmap­ for­ ability­ are­ described­ and­ referenced­ es­available­associated­with­these­as- smart­ grid­ interoperability­ standards,­ in­ the­ NIST­ Interoper- sets­ and­ look­ for­ ways­ Release­ 1.0,­ Office­ of­ the­ National­ ability­ Framework­ Ver- Future research to­ tap­ into­ the­ data­ for­ ­Coordinator­for­Smart­Grid­Interoper- sion­1.0.­Developing­the­ more­comprehensive­di- ability,­National­Institute­of­Standards­ proper­ foundation­ ele- in asset agnostics,­ performance­ and­ Technology,­ NIST­ Special­ Publi- ments­including­security­ assessment,­ lifetime­ es- cation­1108,­Jan.­2010. and­ leveraging­ industry­ management timation,­ maintenance­ J.­ Hughes,­ C.­ Brunner,­ X.­ Yang,­ S.­ standards­ to­ ensure­ optimization,­ and­ re- Zeng,­and­A.­Apostolov,­“IntelliGrid­trans- ­interoperability­ between­ applications placement­ strategies.­ mission­architecture­development,”­EPRI,­ systems­and­devices­will­ within the The­other­area­of­future­ Palo­Alto,­CA,­Rep.­1013833,­2008.­ help­ minimize­ costs­ of­ work­will­need­to­be­in­ National­Energy­Technology­Labo- asset­ management­ func- smart grid enhancing­ the­ compu- ratory,­U.S.­Department­of­Energy­Of- tions­and­many­new­ap- tational­ capabilities­ to­ fice­of­Electricity­Delivery­and­Energy­ plications­that­we­haven’t­ should take deal­ with­ the­ large­ vol- Reliability­ (June­ 2009).­ A­ vision­ for­ even­thought­of­yet. umes­ of­ asset­ specific­ the­ smart­ grid.­ [Online].­ Available:­ place in two data­ and­ developing­ al- http://www.netl.doe.gov/smartgrid/ Research Needs areas. gorithms­ to­ adequately­ referenceshelf/whitepapers/Whitepa- Future­research­in­asset­ interpret­ the­ data­ and­ per_The%20Modern%20Grid%20Vi- management­ applications­ within­ the­ turn­ it­ into­ actionable­ ­information.­ sion_APPROVED_2009_06_18.pdf smart­grid­should­take­place­in­two­ar- Articles­in­this­issue­give­you­an­idea­ GridWise­ Architecture­ Council­ eas.­The­first­should­be­with­respect­to­ of­ progress­ already­ ­being­ made­ in­ (GWAC)­ [Online].­ Available:­ http:// specific­ assets,­ such­ as­ transformers­ these­areas. www.gridwiseac.org/­ p&e

1memberhalf_11pe.inddnovember/december 1 2010 ieee power & energy magazine1/12/11 9:53 AM 63 Reprinted from November/December 2010 issue of Power & Energy magazine

Paul Myrda

in my view my in optimizing assets smart grid-enabled asset management

Power delivery systems face ods, and improved interfaces with the definition of the smart grid several new challenges, including the decision support. it is critical that builds on the work done in ePri’s sheer complexity that results from the the asset management applications intelliGrid [2] program, in the mod- introduction of new devices such as be considered along with a wide va- ern Grid initiative (mGi) [3], and in phasor measurement units (PmUs), riety of other applications as we de- the Gridwise architectural council advanced controllers and sensors on velop the overall requirements for (Gwac) [4]. these considerable ef- equipment throughout the system, in- the smart grid infrastructure. forts have developed and articulated telligent electronic devices (ieds) in the vision statements; architectural Psubstations, smart meters, electric and Smart Grid Background principles; barriers; benefits; tech- hybrid vehicles, photovoltaic genera- the term “smart grid” here refers to a nologies; and applications, policies, tion, distributed storage systems, and modernization of the electricity delivery and frameworks that help define the wind turbines. Power system manage- system so it monitors, smart grid. this back- ment, including managing the assets protects, and automati- ground work was pulled themselves, must be able to deal with cally optimizes the opera- Performance- together into an overall these complexities while at the same tion of its interconnected based asset framework by Nist, time preserving the reliability and elements, from the central and the framework availability of the existing system. and distributed generator management document was pub- while power delivery asset man- through the high-voltage lished last year [1]. agement concepts have been around transmission network and will be a for well over ten years, decision mak- the distribution system, Operations ers have had to settle for systems based to industrial users and powerful tool the system operator on incomplete information and con- building automation sys- that electric is responsible for the strained data collection systems. with tems, to energy storage smooth operation of the advent of the smart grid and its ex- installations, and to end- utilities can use the power system. in tensive communications infrastructure use consumers and their transmission opera- and computational capability, the asset thermostats, electric vehi- in coordination tions, energy manage- manager will be able to determine the cles, appliances, and other ment systems (emss) health and performance of specific as- household devices. with the are used to analyze and sets as a function of system conditions. the smart grid will be fundamental operate the transmission this opens up a new realm of possibil- characterized by a two- power system reliably ity for optimizing asset management. way flow of electricity technologies of and efficiently, while in Performance-based asset man- and information to cre- distribution operations, agement will be a powerful tool that ate an automated, widely the smart grid. similar distribution man- electric utilities can use in coordi- distributed energy deliv- agement systems (dmss) nation with the fundamental tech- ery network. it incorporates into the are being developed and implemented nologies of the smart grid: integrated grid the benefits of distributed com- for analyzing and operating the distri- communications, sensing and mea- puting and communications to deliver bution system. surement, advanced control meth- real-time information and enable the the monitoring of assets within the near-instantaneous balance of supply smart grid will be more cost effective Digital Object Identifier 10.1109/MPE.2010.938199 and demand at the device level. when it can take advantage of a commu- Continued on page 60 64 ieee power & energy magazine 1540-7977/10/$26.00©2010 IEEE november/december 2010 affiliateAdpe.indd 1 1/7/11 12:35 PM Taking the Measure of the Smart Grid around the World