THE SMART GRID:
A PRAGMATIC APPROACH
A “State-of-Play” Discussion Paper Presented by the Canadian Electricity Association

TABLE OF CONTENTS
Executive Summary ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ..2 Introduction. ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ..3 I. Definition and Objectives of the Smart Grid. ... ... ... ... ... ... ... ... ... ... ... ..5 A) Definition.. ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ..5 B) Objectives ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ..5 II. The Smart Grid’s Five Capabilities .. ... ... ... ... ... ... ... ... ... ... ... ... ... ... ..7 A) Demand Response .. ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ..7 B) Facilitation of Distributed Generation ... ... ... ... ... ... ... ... ... ... ... ... ... ... ..7 C) Facilitation of Electric Vehicles . ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ..8 D) Optimization of Asset Use ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ..8 E) Problem Detection and Mitigation ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ..9 III. Building Blocks . ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... A) Hard Infrastructure .. ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... B) Soft Infrastructure ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... C) Summary Map of Building Blocks ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... 11 11 13 16

IV. Growing Pains and Lessons Learned... ... ... ... ... ... ... ... ... ... ... ... ... ... 17 V. An Optimal Path Forward. ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... 20 Conclusion . ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... 22 Sources .. ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... 23

1

EXECUTIVE SUMMARY
The electricity sector has become the focus of heightened policy interest in Canada, as elsewhere, in the context of escalating concerns over emissions, security, and energy demand growth. In this elevated policy context, the smart grid has been much discussed often as a panacea rather than simply the continued maturation of an electricity network that was already on a steady path to automation—and indeed already had some “smart” components.
Unfortunately, this has led to heightened expectations by customers that have yet to be met. As such, the industry nds itself at a crossroads between initial enthusiasm based on industry excitement, and the more pragmatic, cautious path forward. It simply cannot be overemphasized: without customer consent, the deployment of the smart grid will surely stall. To progress towards a smart grid roll-out that is both valuable to stakeholders and widely accepted by customers it is important to understand exactly what a smart grid is. There are numerous de nitions, but the electricity industry in Canada sees the smart grid as a suite of information-based applications made possible by increased automation of the electricity grid, as well as the underlying automation and communication infrastructure itself. As the underpinning to the business case, the various applications and automation technologies must deliver on one or more of the following bene ts: grid resilience, environmental performance, or operational ef ciencies. The transition to a more automated grid—in pursuit of the above mentioned bene ts—entails changes and enhancements across the grid value chain, from how the electricity supplier operates, to how the network is structured, to how the end user interacts with the grid infrastructure. These changes can be organized into ve categories, and constitute the smart grid’s key characteristics or capabilities: demand response, facilitation of distributed generation, facilitation of electric vehicles, optimization of asset use, and problem detection and mitigation. Hard infrastructure, such as smart meters, network devices, energy storage, and smart appliances, as well as soft infrastructure such as interoperability standards, cyber security protocols, the 1.8 Ghz spectrum, and stakeholder engagement, represent the building blocks that support the ve key capabilities. Interwoven into each of these characteristics and building blocks is the theme of improving the customer experience through new service offerings, reduced delivery charges for those offerings, and faster response times. With this understanding of what constitutes a smart grid, it is important to review growing pains and early lessons learned in order to assess how, as a sector, we must adapt to move forward. Security, privacy, implementation cost and stakeholder engagement have each been areas of concern to date; vendors, policy-makers, regulators and utilities must work together to ensure that our collective shareholder, the customer, recognizes the full worth of each installed component of the smart grid. Generally speaking, the business case for automation has been proven time and time again, and the electricity industry will gain value from automation as well. The remaining question marks surround not if, but rather which technologies, at who’s pace, and at what level of public acceptance. This paper provides the basis for discussion of how we will collectively move from “high expectations” through “the valley of despair” and onto “continuous improvement”; this ought to be done through a process of confronting reality, crafting a vision, and communicating belief in the process.

2

INTRODUCTION
To some, the term “smart grid’ has already overstayed its welcome. Governments in some jurisdictions, such as Ontario, are having to defend their mandated smart grid roll-outs to a public that is increasingly wary of smart meters and time-of-use pricing, is pushing back against near-by distributed generation projects, and has been slow to show interest in electric vehicles. In addition, electricity regulators are cautiously examining the costs and benefits to electric utility customers, and utilities are examining whether the smart grid equipment available today is quite as “smart” as it is touted to be. Therefore, despite theoretical benefits and some early demonstrated successes, the smart grid appears to be at a critical phase in its development. A change management consultant with a flair for the dramatic might say that at least some aspects of the smart grid concept have entered “the valley of despair” which often follows “high expectations” and precedes “continuous improvement” on a typical graph showing the phases of change.
HIGH EXPECTATIONS As a concept, the smart grid is intuitive and elegant. Digitization has drastically changed the world of telephony, a mechanic more often wields a diagnostic computer than a wrench, the internet has transformed shopping, and email has replaced the hand-written memo. It was only a matter of time before the electricity grid, recognized by many as the world’s biggest machine, was automated as well.

+
Performance

THE CLASSIC CHANGE CURVE
High expectations

Much better than before

Realization of effort and complexity

X

Light at the end of the tunnel

Time

-

Typical program

“Valley of Despair”

Effective program

3

In addition, and more importantly, the smart grid provides a plausible response to a very important question: how will we, as a society, bring together the elements required to ensure that our energy use is sustainable for future generations. Among other bene ts, the smart grid facilitates the integration of wind and solar and geothermal; it enables a car to be powered by (mostly) hydro power rather than conventional fossil fuels; and it gives customers the knowledge and tools to make the right choices. What’s not to like? This view has been promoted most eagerly by the vendor community that has developed the equipment and software to make it happen. In one recent ad, a scarecrow dances on power lines singing the Wizard of Oz classic “If I Only had a Brain”; similarly, President Obama has been asked to “adopt the goal of giving every household and business access to timely, useful and actionable information on their energy use... [in order to] unleash the forces of innovation in homes and businesses... harness the power of millions of people to reduce greenhouse gas emissions - and save consumers billions of dollars.”1 Now that is marketing. Governments themselves, such as the current Obama administration, have also played a role in the cycle of high expectations, touting smart meters as a way for customers to save money (rather than shift consumption to better optimize generating assets), and assuming that grid advances would allow for the connection of a near unlimited supply of variable generation. Some utilities, as well, underestimated the volume and precision of communication and relationship building required to change a customer base of passive electricity users into active market participants. REALITY CHECK This push-back by the customers is what has led, for some technologies in some jurisdictions, to the rather hyperbolically named “valley of despair”. Questions have arisen about the bene ts of the smart grid by those who pay the rates, and as such

a closer examination of the underlying business case appears to be underway, industry-wide. This is not a new phenomenon; large movements of technological change are almost always over-promised and under-delivered in the rst several years of implementation. An obvious illustrative example is the internet; early expectations were that online shopping would quickly replace bricks-and-mortar stores, a stock bubble formed, expectations were adjusted to the realistic pace of change, and the bubble burst. This is not to say, of course, that online shopping was conceived on a false premise. When the dust settled, the strong online applications remained and a better understanding of the space has led to a process of continuous improvement. The speed at which the utilities and other stakeholders translate lessons learned by the front runners to best practices for all will determine how quickly the smart grid moves towards that nal, steady, upward slope. CONTINUOUS IMPROVEMENT Just as with online shopping, at the heart of the smart grid is a rational concept with real value. Increased automation of the electricity grid will improve its performance and allow for the integration of various applications and usages. That improvement, however, will depend on a myriad of utility-speci c factors including the energy supply mix, the infrastructure already in place, and their relationship with their customers. This paper seeks to explain the broad functionality of the smart grid as it pertains to Canada and the bene ts that it affords to both customers and grid operators, while also setting the stage for all stakeholders to work together for the continuous improvement of the smart grid. Because call it what you would like, the smart grid is moving forward— and that’s a good thing.

1

A Letter to the President of the United States. April 5, 2010. Google Inc. et al.

4

I. DEFINITION AND OBJECTIVES OF THE SMART GRID
A) Definition
The smart grid represents an array of visions to an array of stakeholders. Due to this variance, as well as the complexity of the technologies involved, it is not surprising that the smart grid has given rise to a number of de nitions and explanations. Here are three examples of descriptions recently published by trusted authorities: » “A smart grid is a modern electricity system. It uses sensors, monitoring, communications, automation and computers to improve the exibility, security, reliability, ef ciency, and safety of the electricity system.”2 » “The smart grid takes the existing electricity delivery system and makes it ‘smart’ by linking and applying seamless communications systems that can: gather and store data and convert the data to intelligence; communicate intelligence omnidirectionally among components in the ‘smart’ electricity system; and allow automated control that is responsive to that intelligence.3 » “An automated, widely distributed energy delivery network, the Smart Grid will be characterized by a two-way ow of electricity and information and will be capable of monitoring everything from power plants to customer preferences to individual appliances. It incorporates into the grid the bene ts of distributed computing and communications to deliver real-time information and enable the near-instantaneous balance of supply and demand at the device level.”4 From these de nitions key themes emerge: communication, integration and automation that is sustainable, economic, and secure. Incorporating these themes, this paper offers the following concise de nition of a smart grid: the smart grid is a suite of informationbased applications made possible by increased automation of the electricity grid, as well as the underlying automation itself; this suite of technologies integrates the behaviour and actions of all connected supplies and loads through dispersed communication capabilities to deliver sustainable, economic and secure power supplies. From this de nition, the key objectives of the smart grid come into view.

B) Objectives
Drawing on the above de nition, smart grid investments should support at least one of the following objectives: increase grid resilience, improve environmental performance, or deliver operational ef ciencies including workplace safety. RESILIENCE Grid reliability is non-negotiable. A 2004 study by researchers at the Berkeley National Laboratory found that power interruptions cost the American economy $80 billion per year; other estimates are as high as $150 billion per year.5 Moreover, the North American Electric Reliability Corporation has noted that “reliably integrating high levels of variable resources—wind, solar, ocean and some forms of hydro—into the North American bulk power system will require signi cant changes to the traditional methods used for system planning and operations.”6 Proponents claim that the smart grid will facilitate these changes by enabling additional dispersed supply and by enhancing corrective capabilities where problems occur. While the smart grid may indeed enhance security in some aspects, however, the additional information technology of the smart grid may also render it more vulnerable than the conventional grid to cyber attacks, and as such may pose a very real threat to reliability.

2

Paul Murphy et. al., Enabling Tomorrow’s Electricity System: Report of the Ontario Smart Grid Forum, http://www.ieso.ca/imoweb/pubs/ smart_grid/Smart_Grid_Forum-Report.pdf (September, 2010) Miles Keogh, The Smart Grid: Frequently Asked Questions for State Commissions, The National Association of Regulatory Utility Commissioners, May 2009, p. 2, http://www.naruc.org/Publications/NARUC%20Smart%20Grid%20Factsheet%205_09.pdf, (June, 2010) The Smart Grid: An Introduction, U.S. Department of Energy, http://www.oe.energy.gov/DocumentsandMedia/DOE_SG_Book_Single_ Pages(1).pdf (September, 2010) Kristina Hamachi LaCommare and Joseph H. Eto, Understanding the Cost of Power Interruptions to U.S. Electricity Consumers, Ernest Orlando Lawrence Berkeley National Laboratory, September 2004, e.g., Figure ES-1 among other discussions in the paper: http://certs. lbl.gov/pdf/55718.pdf (September 2010). Accommodating High Levels of Variable Generation, special report of the North American Electric Reliability Corporation, Princeton, New Jersey, April 2009, Executive Summary, http://www.nerc.com/ les/IVGTF_Report_041609.pdf (August, 2010).

3

4

5

6

5

ENVIRONMENTAL PERFORMANCE Politicians, environmental stakeholders and the general public are increasingly looking to the electricity sector to reduce the emissions resulting from power generation as well as to drive further emission reductions by replacing liquid fossil fuels in the transportation sector. The smart grid is expected to drive carbon emissions reductions by facilitating renewable power generation, enabling electric vehicles as replacements for conventional vehicles, reducing energy use by customers, and reducing energy losses within the grid. Each of these positive outcomes requires vital information to be available to the grid operators that has not traditionally been available; distribution automation furnishes these required tools. OPERATIONAL EFFICIENCIES The smart grid will be expensive to develop and deploy, but if implemented pragmatically should provide operational ef ciencies that outweigh these costs. The electricity industry went through a growth phase in the 1970’s and 1980’s, and aging infrastructure is coming due for replacement. In fact, the electricity industry in Canada is expected to invest $11 billion in infrastructure replacement in each of the next 20 years just to replace existing assets. This is a cost that must be incurred with or without the automation of the grid. Rather than replacing assets with identical assets, however, the smart grid, if planned pragmatically, represents the technological upgrades that will pay a positive return on the investment over the deployed life cycle through energy demand reductions, savings in overall system and reserve margin costs, lower maintenance and servicing costs (e.g. reduced manual inspection of meters), and reduced grid losses, and new customer service offerings. While some bene ts to operational ef ciency t quite nicely into a business plan, such as line loss reduction or improved asset management, some elements rely on a societal assessment of worth, rather than an accountant’s calculation of “value”. For example, new subdivisions since the 1960s have been built with a preference for hiding distribution wires underground. While this practice provides tangible bene ts that can be measured (i.e. extending the life of wires because they are not exposed to the elements), the business case is also supported by intangible bene ts (i.e. the aesthetics value of

not seeing the distribution system running through the neighbourhood). This concept of tangible versus intangible operational ef ciencies can also be illustrated through workplace safety, a topic that Canadian utilities take very seriously. This commitment to safe work environments is supported by several functionalities available through the smart grid, notably by reducing time on the road for meter reading, alerting workers of islanding, and allowing for some grid repairs to be performed remotely. Avoiding injuries certainly provides tangible operational bene ts such as reducing lost time due to injury, but a portion of the bene t is attributed to the intangible health and safety bene ts accrued to any worker whose job is made safer. Operational improvements such as these are tough to quantify in a business case, but like undergrounding, once their worth is proven, rather than simply their “value”, they are likely to become the new industry standard.

6

II. THE SMART GRID’S FIVE CAPABILITIES
The transition to a more automated grid—in pursuit of environmental, efficiency and resilience benefits— entails changes and enhancements across the grid value chain, from how the electricity supplier operates, to how the network is structured, to how the end user interacts with the grid infrastructure. These changes can be organized into five broad categories, and constitute the smart grid’s key characteristics or “capabilities”.
A) Demand Response
This capability refers to the capacity of the user or operator to adjust the demand for electricity at a given moment, using real-time data. Demand response can take the form of active customer behaviour in response to various signals, generally the price of electricity at the meter, or it can be automated through the integration of smart appliances and customer devices which respond to signals sent from the utility based on system stability and load parameters. For example, a residential hot water heater could be turned off by a utility experiencing high electricity loads on a hot day, or could be programmed by its owner to only turn on at off-peak times. Active demand management can help smooth load curves, which in turn can reduce the required reserve margins maintained by electricity generators. Some pilot projects can already claim results in this respect: the Olympic Peninsula Project, overseen by the Paci c Northwest National Laboratory on behalf of the US Department of Energy, dropped peak power usage by 15 percent. A similar project from Constellation Energy in Baltimore, Maryland, cut peak power demand by at least 22 percent—and as much as 37 percent. 7 These capabilities have been rolled out in several Canadian jurisdictions to date; however the value of this technology depends on a number of factors. The rst, of course, is customer take-up. If electricity customers do not sign up for voluntary utility load control programs or do not purchase the smart appliances and devices required, demand response programs will have little effect. Additionally, if the generating mix in a particular jurisdiction allows it to economically adapt to electricity demand, the value of demand response programs is diminished. In Alberta, for example, the average power divided by the peak power output, or “load factor”, for the province is about 80%, which is quite high. As such, the value of peak shaving programs is diminished as compared to other Canadian jurisdictions with load factors below 80%. It is important to note that demand response and energy conservation are not one and the same. Successful demand response smoothes out consumption levels over a 24-hour period, but does not encourage decreased consumption. Smart grid technologies that promote a reduction in the use of electricity include the Advanced Metering Infrastructure (AMI) and the Home Area Network (AM), both of which allow for increased customer control over their energy use.

B) Facilitation of Distributed Generation
As demand response is the management of system outputs, the facilitation of distributed generation is the management of system inputs. Some in the industry refer to the combined optimal management of both to be the “achievement of ow balance.” Traditionally, the grid has been a centralized system with one way electron ows from the generator, along transmission wires, to distribution wires, to end customers. One component of the smart grid allows for both movement and measurement in both directions, allowing small localized generators to push their unused locally generated power back to the grid and also to get accurately paid for it. The wind and the sun, however, generate energy according to their own schedule, not the needs of the system. The smart grid is meant to manage intermittency of renewable generation through advanced and localized monitoring, dispatch and storage. In Ontario, the Energy Board has directed that it is the responsibility of the generator to mitigate any negative effects that connected supply may have

7

David Biello, The Start-Up Pains of a Smarter Electricity Grid, Scienti c American, May 10, 2010, http://www.scienti camerican.com/article.cfm?id=start-up-pains-of-smart-grid (September 2010).

7

on the distribution grid in terms of voltage variances and power quality. The optimal solution set to accomplish this, however, is still being examined. In addition to intermittency challenges, distributed generation can cause instances of “islanding” in which sections of the grid are electri ed even though electricity from the utility is not present. Islanding can be very dangerous for utility workers who may not know that certain wires have remained live during a power outage. Ideally, real time information will allow islanded customers to remain in service, while posing no risk to utility workers. Again, the automation afforded by the smart grid offers a means to this end. When Louisiana was hit by Hurricane Gustav on September 1, 2008, an island was formed of about 225,000 customers who were disconnected from the main electricity grid. According to Entergy, the responsible utility, “synchrophasors installed on key buses within the Entergy system provided the information needed for the operators to keep the system operating reliably.”8 This technology saved the utility an estimated $2-$3 million in restoration costs, and kept all customers in service (thereby avoiding economic losses to regional businesses).9

example of the potential risk to utilities of getting caught in the middle. Many policy makers and car manufacturers correctly point out that widespread charging infrastructure may help incent customers to switch to electric vehicles. While this is true, we must recognize that charging infrastructure alone may not be enough to change customer behaviour; until a breakthrough technology is discovered by the automotive industry, electric vehicles will still have relatively high price tags and limited range. As such, prudence dictates that utility investments in EV infrastructure ought to respond to the automotive purchasing patterns of their customers rather than laying the groundwork for a fuel switch that is still largely dependent on technological breakthroughs. If utilities invest in infrastructure now, and the EV market takes longer than promised to develop, customers may not feel well served.

D) Optimization of Asset Use
Monitoring throughout the full system has the potential to reduce energy losses, improve dispatch, enhance stability, and extend infrastructure lifespan. For example, monitoring enables timely maintenance, more ef cient matching of supply and demand from economic, operational and environmental perspectives, and overload detection of transformers and conductors. Or as Miles Keogh, Director of Grants and Research at the National Association of Regulatory Utility Commissioners in the US, argues in a recent paper, system optimization can occur “through transformer and conductor overload detection, volt/var control, phase balancing, abnormal switch identi cation, and a host of ways to improve peak load management.” Thus, as he concludes, “while the smart meter may have become the ‘poster child’ for the smart grid, advanced sensors, synchro-phasors, and distribution automation systems are examples of equipment that are likely to be even more important in harnessing the value of smart grid.”11

C) Facilitation of Electric Vehicles
The smart grid can enable other bene cial technologies as well. Most notably, it can support advanced loading and pricing schemes for fuelling electric vehicles (EVs). Advanced Metering Infrastructure would allow customers to recharge at off-peak hours based on expected prices and car use patterns, while bidirectional metering could create the option for selling back stored power during on-peak hours. Although signi cant EV penetration is still a medium to long-term projection, some cities and regions have started experiments and the existence of a smart grid is essential to their uptake. This area of the smart grid provides an illustrative

8

Floyd Galvin and Chuck Wells, “Detecting and Managing the Electrical Island Created by Hurricane Gustav,” Success Stories, North American Synchrophasor Initiative, p. 1, http://www.naspi.org/stories/pilot_fundamental/entergy_hurricane_gustav.pdf, (July, 2010). Galvin and Wells, 2. Miles Keogh, “The Smart Grid: Frequently Asked Questions for State Commissions,” The National Association of Regulatory Utility Commissioners, May 2009. . [con rm citation]. Keogh, 4.

9 10

11

8

For example, smart grid monitoring helps utilities asses their line proximity issues as it relates to trees and tree growth, because dense growth results in a signi cant increase in the number of short voltage blips that occur. Early detection of these short line contacts by trees will assist utilities in their “just in time” tree programs, effectively focussing crews on the correct “problem areas”. In addition, network enhancements, and in particular improved visualization and monitoring, will enable “operators to observe the voltage and current waveforms of the bulk power system at very high levels of detail.” This capability will in turn “provide deeper insight into the real-time stability of the power system, and the effects of generator dispatch and operation;” and thereby enable operators to “optimize individual generators, and groups of generators, to improve grid stability during conditions of high system stress.”12

what they must do to restore service. It is not rare, in fact, for a utility customer care representative to ask a caller to step outside to visually survey the extent of the power loss in their neighbourhood. It is a testament to the high levels of reliability enjoyed by electric utility customers that most have never experienced this; however, it is also evidence of an antiquated system. While SCADA and other energy management systems have long been used to monitor transmission systems, visibility into the distribution system has been limited. As the grid is increasingly asked to deliver the above four capabilities, however, dispatchers will require a real-time model of the distribution network capable of delivering three things: 1) real-time monitoring (of voltage, currents, critical infrastructure) and reaction (re ning response to monitored events); 2) anticipation (or what some industry specialists call “fast look-ahead simulation”); and 3) isolation where failures do occur (to prevent cascades). On any given day in the United States, roughly “500,000 U.S. customers are without power for two hours or more”13 costing the American economy between $70 and $150 billion a year.14 This signi cant impact on economic activity provides a strong incentive to develop the smart grid, which is expected to reduce small outages through improved problem detection and isolation, as well as storage integration. It is also

E) Problem Detection and Mitigation
Many utility customers do not realize the limited information currently available to grid operators, especially at the distribution level. When a blackout occurs, for example, customer calls are mapped to de ne the geographic area affected. This, in turn, allows utility engineers to determine which lines, transformers and switches are likely involved, and

12

Methodological Approach for Estimating the Benefits and Costs of Smart Grid Demonstration Projects, EPRI, January, 2010, p. 4–21 (July, 2010). Massoud Amin and Phillip F. Schewe, “Preventing Blackouts: Building a Smarter Power Grid,” Scientific American, August 13, 2008, http://www.scienti camerican.com/article.cfm?id=preventing-blackouts-power-grid&page=3, (September, 2010) Scienti c American says that “estimates peg the economic loss from all U.S. outages at $70 to $120 billion a year,” while NARUC says “outages cost between $80 and $150 billion every year.”

13

14

9

expected to reduce the likelihood of big blackouts, such as the infamous 2003 blackout that impacted most of the Eastern seaboard. The 2003 blackout left more than 50 million people without power for up to two days, at an estimated cost of $6 billion, and contributed to at least 11 deaths.15 A root cause analysis revealed that the crisis could not have begun in a more innocuous way: a power line hit some tree branches in northern Ohio. An alarm failed to sound in the local utility, other lines also brushed against trees, and before long there was a cascade effect—a domino of failures—across eight US states and one Canadian province. With proper monitoring, now capable through smart grid innovations, some proponents believe that a cascading blackout mirroring that of 2003 should become so remote a possibility as to become almost inconceivable.16 Intelligent monitoring on a smarter grid allows for early and localized detection of problems so that individual events can be isolated, and mitigating measures introduced, to minimize the impact on the rest of the system. The current system of supervisory control and data acquisition (SCADA), much of it developed decades ago, has done a reasonably good job of monitoring and response. But it has its limits: it does not sense or monitor enough of the grid; the process of coordination among utilities in the event of an emergency is extremely sluggish; and utilities often use incompatible control protocols—i.e. their protocols are not interoperable—with those of their neighbours. If Ohio already had a smart grid in August 2003, history might have taken a different course.17 To begin with, according to Massoud Amin and Phillip Schewe in a Scienti c American article, “fault anticipators… would have detected abnormal signals and redirected the power… to isolate the disturbance several hours before the line would have failed.”18 Similarly, “look-ahead simulators

would have identi ed the line as having a higherthan-normal probability of failure, and self-conscious software… would have run failure scenarios to determine the ideal corrective response.” As a result, operators would have implemented corrective actions. And there would be further defences: “If the line somehow failed later anyway, the sensor network would have detected the voltage uctuation and communicated it to processors at nearby substations. The processors would have rerouted power through other parts of the grid.” In short: customers would have seen nothing more than “a brief icker of the lights. Many would not have been aware of any problem at all.”19 Utility operators stress that the smart grid does not spell the end of power failures; under certain circumstances such as these, however, any mitigation could prove very valuable indeed. A more reliable grid is also a safer grid. First, as discussed previously, smart grid technology allows for “anti-islanding” when needed. Detection technology can ensure that distributed generators detect islanding and immediately stop producing power. Second, power failures can leave vulnerable segments of the population, such as the sick or elderly, exposed to the elements or without power required by vital medical equipment. Third, safety is also enhanced through electricity theft reductions. As BC Hydro points out, “energy diversions pose a major safety risk to employees and the public through the threat of violence, re and electrocution.” 20

15

JR Minkel, “The 2003 Northeast Blackout – Five Years Later,” Scientific American, August13, 2008, http://www.scienti camerican.com/ article.cfm?id=2003-blackout- ve-years-later, (August, 2010). Amin and Schewe. Amin and Schewe. Amin and Schewe BC Hydro, http://www.bchydro.com/planning_regulatory/projects/smart_metering_infrastructure_program/program_overview_and_ status.html (October, 2010)

16 17 18 19

10

III. BUILDING BLOCKS
The five capabilities just reviewed— demand response, facilitation of distributed generation, facilitation of electric vehicles, optimization of asset use, and problem detection and mitigation—have excited considerable interest in policy discussions about the smart grid. To assess the merits of each, however, we ought to bear in mind that their value is derived from their ability to contribute towards the three ultimate objectives of increased resilience, improved environmental performance, and operational efficiencies. In other words, we need to consider their contribution in practical terms.
This question of practicality gives rise to a consideration of the building blocks needed to implement the various capabilities. Implementation of a smart grid will require investments and changes in tangible infrastructure complemented by investments and changes in soft infrastructure. A detailed understanding of the bene ts and challenges for both of these categories is required when assessing the business case for the various capabilities of the smart grid. and also likely the most expensive, building block supporting a smart grid. As of September 30, 2009, electricity distributors in Ontario had installed approximately 2,883,000 residential and 171,000 general service (; March 23, 2010. NIST Framework and Roadmap for Smart Grid Interoperability Standards, Release 1.0, Of ce of the National Coordinator for Smart Grid Interoperability, National Institute of Standards and Technology, U.S. Department of Commerce, NIST Special Publication 1108, January, 2010. “Pathways to a Low-Carbon Economy: Version 2 of the global greenhouse gas abatement cost curve,” McKinsey & Company, January 2009: https://solutions.mckinsey.com/climatedesk/ “PG&E smart meter problem a PR nightmare,” November 21, 2009, http://www.smartmeters.com/the-news/690pgae-smart-meter-problem-a-pr-nightmare.html “Summary of Smart Grid Bene ts and Issues,” Illinois Smart Grid Initiative, http://www.cnt.org/news/media/isgi-summaryof-bene ts-and-issues-6-08.pdf Raftery, Tom. “PG&E smart meter communication failure—lessons for the rest of us,” Green Monk blog, December 16, 2009, http://greenmonk.net/pge-smart-metercommunication-failure/

Ramsay, James Bradford. “Implementation of Smart Grid Technology,” Initial Comments of the National Association of Regulatory Utility Commissioners in Response to NPB Public Notice #2, Before the Federal Communications Commission, October 2, 2009, DA 09-2017, http://www.naruc.org/Testimony/09%20 1002%20NARUC%20Smart%20Grid%20 comments. n.pdf “Reasons for Decision to Order G-168-08,” In the Matter of FortisBC Inc. and an Application for a Certi cate of Public Convenience and Necessity for its Advanced Metering Infrastructure, December 3, 2008, http://www.bcuc.com/Documents/ Proceedings/2008/DOC_20449_G-168-08_ with-Reasons-for-Decision.pdf “Sector Smart Meter Audit Review Report,” Ontario Energy Board Regulatory Audit and Accounting, March 31, 2010: http://www.oeb.gov.on.ca/OEB/_Documents/ Audit/Smart_Meter_Audit_Review_Report.pdf> Smart Grid—Technology Innovation Group Report, May 26, 2010, Tokyo Japan. “Smart networks position paper,” Energy Networks Association, September 2009. Smart Privacy for the Smart Grid: Embedding Privacy into the Design of Electricity Conservation, PbD, Information and Privacy Commissioner, Toronto, Ontario, November 2009, http://www.ipc.on.ca/images/resources/ pbd-smartpriv-smartgrid.pdf Smith, Rebecca. “Smart Meter, Dumb Idea?” Wall Street Journal, April 27, 2009, http://online.wsj.com/article/ SB124050416142448555.html “Canadian Smart Grid Framework,” Canadian Electricity Assocation, March 25, 2010.

24