Getting Smart

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

I
MASTER SERIES

ITS 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 specically constitutes a smart grid, a consensus is forming regarding its general attributes. The 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 consumers in demand response. It operates resiliently against both physical and cyber attacks. It provides quality power that meets 21stcentury needs. It accommodates all generation and storage options.

Digital Object Identier 10.1109/MPE.2009.935557

By Enrique Santacana, Gary Rackliffe, Le Tang, and Xiaoming Feng

1540-7977/10/$26.002010 IEEE
IEEE power & energy magazine

march/april 2010

41

 It enables new products, services, and markets. It optimizes asset utilization and operating efciency.

In Europe, a smart grid is described, according to a recent European Commission report, as one that is exible as it fullls customers needs while responding to the changes and challenges ahead accessible as it grants connection access to all network users, particularly renewable power sources and high-efciency local generation with no or low carbon emissions reliable as it assures and improves security and quality of supply, consistent with the demands of the digital age, with excellent resilience in the face of hazards and uncertainties economical as it provides best value through innovation, efcient energy management, and level playing eld competition and regulation. China is also developing the smart grid concept. The term smart grid refers to an electricity transmission and distribution system that incorporates elements of traditional and cutting-edge power engineering, sophisticated sensing and monitoring technology, information technology, and communications to provide better grid performance and to support a wide range of additional services to consumers. A smart grid is not dened by what technologies it incorporates but rather by what it can do, according to the nonprot Joint US-China Cooperation on Clean Energy. Such high-level characterization of the smart grid, while helpful at the strategic level, leaves plenty of room for confusion and very different interpretations on the part of laymen and professionals alike, due to a lack of specics. You are not alone in wondering, Exactly what is the smart grid? At the National Governors Association convention in February 2009, the CEO of a major utility began his speech with the confession that he didnt really know what smart grid meant (see Why the Smart Grid Industry Cant Talk the Talk, in the For Further Reading section). Its little wonder that people sometimes confuse the smart grid with smart meters and advanced metering infrastructure (AMI) or with interoperability in communications. In this article we offer our perspective on the smart grid. We will look at the drivers for the smart grid, sketch out its scope, discuss what makes the smart grid smart, and envision its distinguishing features. We will focus on the technical challenges that a smart grid must deal with and postulate the capabilities a smart grid must have in order to meet those challenges successfully. Due to limitations of space, we will not delve into the details of how to meet such challenges, but we will sketch a potential example in order to give technically inclined readers an idea of the technologies that are emerging on the horizon.

Why Do We Need the Smart Grid?

An electric grid consists of three main subsystems: the generation sources (various power plants); the delivery system
42
IEEE power & energy magazine

(transmission and distribution networks); and the end customers (residents, commercial buildings, industrial installations, and others). The electric grid is unique in that electrical supply and demand must remain tightly balanced at all times, since for most of the history of the electric grid there has been no commercial solution for large-scale storage of electricity to compensate for any excess or shortfall in power. In the past, this balancing act was performed by the vertically integrated utilities that controlled both the generation and the delivery systems. Power grids in the industrialized countries are aging and being stressed by operational scenarios and challenges never envisioned when the majority of them were developed many decades ago. The main challenges are summarized below. Deregulation unleashed unprecedented energy trading across regional power grids, presenting power ow scenarios and uncertainties the system was not designed to handle. The increasing penetration of renewable energy in the system further increases the uncertainty in supply and at the same time adds stress to the existing infrastructure due to the remoteness of the geographic locations where the power is generated. Our digital society depends on and demands a power supply of high quality and high availability. The threat of terrorist attacks on either the physical or the cyber assets of the power grid introduces further uncertainty. There is an acute need to achieve sustainable growth and minimize environmental impact via energy conservation, i.e., by switching to green and renewable energy sources. We can only meet this objective by increasing energy efciency, reducing peak demand, and maximizing the use of renewable energy. The growing consensus in the industry and among many national governments is that smart grid technology is the answer to these challenges. This trend is evidenced by the specic provision and appropriation of multi-billiondollar amounts on the part of the U.S. government (in the 2009 stimulus program, for example) for research and development, demonstration, and deployment of smart grid technologies and the associated standards. The European Union and China have also announced huge levels of funding for smart grid technology research, demonstration, and deployment. The objective of transforming the current power grid into a smart grid is to provide reliable, high-quality electric power to digital societies in an environmentally friendly and sustainable way. This objective will be achieved through the application of a combination of existing and emerging technologies for energy efciency, renewable energy integration, demand response, wide-area monitoring and control, selfhealing, HVDC, exible ac transmission systems (FACTS), and so on.
march/april 2010

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 posiand 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 design and operational paradigm of the grid: from central to distributed resources, from predictable power ow directions to unpredictable directions, from a passive grid to an active grid. In short, the grid will be more dynamic in its conguration 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 question, lets 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 function 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 the car is on a level road, increases the engine output when the car is going uphill, and reduces engine output when the car is going downhill. All this is done automatically, without the intervention of the
march/april 2010 Traditional Grid Smart Grid

Traditional Grid Centralized Generation Power Flows Downhill Utility Controls Connections Behavior: Predictable

Smart Grid Generation Everywhere Power Flows from Everywhere Anyone May Participate Behavior: Chaotic

figure 2. Transition from the present grid to a smart grid.

IEEE power & energy magazine

43

 Vehicle Speed Steering Wheel Controls Clutch Pedal Brake Pedal

Cruise Control Computer

Vacuum Valve Control

Vacuum Actuator Cable-to-Throttle Valve Throttle Position Throttle Valve

figure 3. Schematic diagram of vehicle cruise control.

generate control signals for the vacuum actuator. The vacuum actuator increases or reduces the throttle valve by means of a cable. The changes in the throttle valve position change the engine power output and in turn the speed of the car. We can identify four essential building blocks necessary in this system: a sensor system to measure system states (automobile speed, brake and pedal position, throttle position) communication infrastructure (wires to collect sensor information and propagate control signals) control algorithms (also known as applications that digest the information and generate control signals intended to change the state of the system) actuators that effect desired changes in the physical system (in this example, the throttle valve position and the engine power output). All four building blocks are needed for this smart function to work. The sensor system and the communication infrastructure let the driver know what is going on, i.e., what speed the car is

going and whether it is accelerating or decelerating. The control algorithms (applications) are where the smartness of the system lies. The control algorithms make intelligent decisions based on the information provided by the communication infrastructure, the available controls, and the desired control objective (maintaining constant speed). But without the actuator system, we can only observe the system passively and helplessly, for the 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. The four basic building blocks identied 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 on the interoperability of the communication and information model, as suggested by the National Institute of Standards and Technology (NIST) Smart Grid Interoperability standard road map and the International Electrotechnical Commission (IEC) documents on smart grid standardization. In Understanding the Smart Grid from Denition to Deployment, the Edison Electric Institute rightly suggested that advanced controls provide the smart in smart grids. To borrow a phrase from the real estate business, the three value generators for the smart grid are application, application, and application. To enable smart applications, we need not only good business logic, control, and optimization theory, we also need new hardware components that can control power ows in the network, as well as the output and the consumption of power.

What Will the Smart Grid Be Like?

To average consumers, the smart grid, for the most part, will remain under the hood, working silently and invisibly. Some interfaces will be exposed to consumers, such as the prototype iPhone interface by which users will be able to check the current electricity price and electricity consumption and remotely turn home appliances on and off. Though fascinating, such technologies represent only the tip of the iceberg. The really important and advanced technologies of the smart grid will remain unnoticed by the general public. When we look beyond the horizon, we envision some salient features of the smart grid that set it apart from the traditional power grid. It is clear that as the systems supply

table 1. The four building blocks of vehicle cruise control mapped to the electric power system. Building Blocks Sensor system Communication infrastructure Control algorithms (applications) Power System Mapping Current transformer (CT), voltage transformer (VT), phasor measurement unit (PMU), smart meter, temperature, pressure, acoustic, and so on Power line carrier (PLC), wireless radio, advanced metering infrastructure (AMI), home area network (HAN), fiber-optic networks 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 HVDC, FACTS, DG, energy storage systems, reclosers, automatic switches, breakers, switchable shunts, on-load tap changers, hybrid transformers, and so on

Actuator system/physical system

44

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 distributed, the systems condition will become more dynamic and less predictable. The development of demand, supply, and power ow control technologies will thus become essential in protecting, managing, and optimizing the new grid. In the following sections we summarize certain other features of the future smart grid.

Distribution of Production
DG (from solar, fuel cell, small wind turbine, and other sources) and energy storage (battery, thermal, and hydrogen) are everywhere in a smart grid. They are not marginal players but highly inuential and integrated parts of the energy web. They provide energy diversity, reduce demand for central fossil-fuel power plants, and increase supply redundancy and system reliability. The distribution of energy production from renewable sources also increases the resilience of the grid in the face of widespread disturbances (e.g., blackouts).

Tightly Integrated Renewable Energy

In the smart grid, energy from diverse sources is combined to serve customer needs while minimizing the impact on the environment and maximizing sustainability. In addition to nuclear-, coal-, hydroelectric-, oil-, and gas-based generation, energy will come from solar, wind, biomass, tidal, and other renewable sources. The smart grid will support not only centralized, large-scale power plants and energy farms but residential-scale dispersed distributed energy sources. These renewable and green sources will be seamlessly integrated into the main grid.

A New Level of Controllability

In the future smart grid, a new generation of power transport and control technologies will have become mature and widely adopted. Current-limiting and current-breaking devices based on solid-state technology will help protect valuable grid assets and isolate faults. Power electronics based transformers will be common. FACTS technology will enable system operators to route power ows along the most efcient paths and nd the best power production mixes and schedules. Advanced applications in the control center will continuously check the state of the grid and determine the best control strategies from among billions of possibilities in real time.

Proliferation of Energy Storage

A smart grid has numerous energy storage centers, large and small, stationary and mobile, that it can use to buffer the impact of sudden load changes and uctuations in wind and solar generation, as well as to shift energy consumption away from peak hours by providing energy balancing, load following, and dynamic compensation of both reactive and real power. The recent development of quick-response battery energy storage systems (BESSs) with voltage source converters (VSCs) has demonstrated the promise and potential benets of energy storage.

Real-Time Grid Awareness

Massively deployed sensors will continuously collect enduser energy consumption data, weather data, and equipment condition and operational status and perform real-time rating in the context of actual distribution and transmission line ows. The information will be disseminated through highly available, exible, open (but secure) two-way communication infrastructures to any point in the grid where it can be used to monitor the status of the grid, predict what will happen next, and develop optimal control strategies.

Growing Mobile Loads and Resources

Many loads and resources connected to the future smart grid will no longer be stationary. Breakthroughs in battery technology are making plug-in electric vehicles (EVs) commercially viable. At all times of day, tens of millions of EVs will be connected to the future grid at parking lots near homes, workplaces, and shopping malls. These EVs will represent both mobile loads and potential sources of power. The battery systems in these vehicles will be charged or discharged via sophisticated coordination protocols in order to smooth out uctuations in power demand in different parts of the grid, avoid power transmission bottlenecks, and render the grid more stable. Controllers will be able to respond to power system condition signals such as voltages and frequencies as well as market signals such as real-time electricity prices.
march/april 2010

The Smart Prosumer and the Grid-Friendly Appliance

End-user equipment will no longer consist of dumb devices but will form interactive and intelligent nodes on the smart grid. End-user energy management systems will monitor the energy consumption situation in residences, ofce buildings, and shopping malls. They will know the consumption patterns and preferences of the occupants, as well as real-time conditions (e.g., market prices, grid stress). They will use the collected information to autonomously interact with the grid to determine the charging and discharging cycles of plug-in electric vehicles,
IEEE power & energy magazine

45

schedule washer and dryer cycles, and optimize HVAC operations without sacricing occupants comfort. Appliances will continuously monitor voltages and frequencies. When the system experiences distress due to unforeseen disturbances, the appliances will modulate the power consumed to reduce the stress on the system and help prevent service disruptions.

The Resurgence of DC
Advancements in materials, power electronics, and sensor technologies will transform the design and operation paradigm of the smart grid. At the generation, transmission, and distribution levels, ac and dc technologies will work together harmoniously. HVDC networks embedded in ac networks will power the worlds megacities but will use only a fraction of the land required for transmission a generation ago. HVDC transmissions will link clean and renewable power at remote or offshore generation sites to the main power grid. Distribution buses in ofce and residential buildings will supply dc power to digital appliances without the need for power adapters. Hybrid grid (ac/dc) architectures for distribution systems will make the grid more exible and reliable.

Real-Time Distributed Intelligence

In the smart grid, advanced grid-monitoring, optimization, and control applications track the operating conditions of grid assets, calculate their ratings, and dynamically balance load and resources to maximize energy delivery efciency and security in real time. The increased interactivity among producers and consumers will mean demand is dynamic rather than static; the grids operating environment will appear chaotic, and power ow directions will change in response to market conditions. A new generation of protection and control technologies will be called on to maintain the safety and security of both the system and its personnel.

The Four Technology Layers of the Smart Grid

The four essential building blocks of the smart grid can be depicted using a layered diagram, as shown in Figure 4. An analogy can be drawn between these layers and those that make up the human body. The bottom layer is analogous

Decision Intelligence

Communication

Sensor/Actuator Power Conversion/Transport/Storage/ Consumption

figure 4. Smart grid technology layers.

46
IEEE power & energy magazine

to the bodys muscles; the sensor/actuator layer corresponds to the bodys sensory and motor nerves, which perceive the environment and control the muscles; the communication layer corresponds to the nerves that transmit perception and motor signals; and the decision intelligence layer corresponds to the human brain. The smartness of the smart grid lies in the decision intelligence layer, which is made up of all the computer programs that run in relays, intelligent electronic devices (IEDs), substation automation systems, control centers, and enterprise back ofces. These programs process the information collected from the sensors or disseminated from the communication and IT systems; they then provide control directives or support business process decisions that manifest themselves through the physical layer. Some application examples are given below: microgrid control and scheduling (demand response and efciency) intrusion detection and countermeasures (cybersecurity) equipment monitoring and diagnostic systems (asset management) wide-area monitoring, protection, and control online system event identication and alarming (safety and reliability) power oscillation monitoring and damping (stability) voltage and var optimization (energy efciency and demand reduction) voltage collapse vulnerability detection (security) autonomous outage detection and restoration (selfhealing) intelligent load balancing and feeder reconguration (energy efciency) self-setting and adaptive relays (protection) end-user energy management systems (consumer participation, efciency) dynamic power compensation, using energy storage and voltage source inverters (efciency and stability). For the decision intelligence layer to work, data (information) need to be propagated from the devices connected to the grid to the controllers that process the information 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 is needed by the decision intelligence layer. In most cases, this means that data are transferred from eld devices back to the utility control center, which acts as the main repository for all the utilitys data. Device-to-device (e.g., controllerto-controller or IED-to-IED) communication, however, is also common, as some real-time functionality can only be achieved through interdevice communication. Interoperability and security are essential to assure ubiquitous communication between systems of different media and topologies and to support plug-and-play for devices that can be autocongured when they are connected to the grid, without human
march/april 2010

intervention. The accelerating deployment of AMI around the world is a big step in building a two-way communication platform for enabling demand response and other advanced distribution applications. The physical layer is where the energy is converted, transmitted, stored, and consumed. Solid-state technology, power electronicsbased building blocks, superconducting materials, new battery technologies, and so on all provide fertile ground for innovations.

Example of a New Controllable Component

On the journey towards the smart grid, there will be many technology breakthroughs that will have game-changing effects on its evolution. What follows is one plausible newtechnology scenario, described here as an illustration of one of the many potential smart grid technologies that could change the grids design and operational philosophy in fundamental ways.

system architecture can be envisioned, like the one shown in Figure 6. The new architecture requires the introduction of new building blocks. We shall refer to the new building block technology as the smart integration module (SIM). SIMs will have the following functionality: connection to the grid (feeder) ac bus for ac loads dc bus for dc loads and connection to energy storage and distributed generation voltage regulation in steady state and in transient fast real and reactive power compensation fault detection and fault current limiting and isolation autonomous distributed intelligent control for shorttime-scale control coordination and optimization for longer-time-scale control.

New Concept of Smart Integration Module

Advancements in power electronics design and fabrication technology make it possible in principle to design smart integration modules (SIMs) with the aforementioned functionality at a competitive cost. If shown to be technologically and commercially viable, SIM technology could change the design and operation philosophy and practice in the distribution and transmission systems in profound ways.

Traditional Grid Design and Premises

The design and operation of the traditional power grid is limited by the basic network components (lines, switches, transformers with on-load tap changers, switchable capacitors) available. The traditional grid is built on the following ve premises: 1) The components are predominantly dumb conductors and are not controllable. 2) Even if they are controllable, they cannot react quickly enough. 3) There is no energy storage; an interruption on the transmission or distribution grid means an interruption of service. 4) Customer demands are not controllable, and the grid can only react passively to the change in demands with centralized control. 5) The grid can only react to the changes by continuously balancing the output of the central power plants in order to remain in a dynamic equilibrium. The lack of energy storage, fast reactive and real power regulation, and distributed generation leads to the traditional 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 within acceptable range at the service point. The main function of the switchable capacitor banks is to provide reactive power close to the loads and reduce reactive power ow on the feeder and energy losses.

Benefits and Impacts

Multiple benets could result from a SIM-based architecture. Energy loss would be reduced to a minimum due to

Voltage Regulator SS Cap ac Bus Transformer Load Transformer Load Load ac Bus

figure 5. Traditional distribution system architecture.

SS Load SIM Energy Storage Load Load Load ac Bus dc Bus DG Load

New Architecture and Enabling Technology

With distributed generation, energy storage, and fastacting converter/inverter technology, a new distribution
march/april 2010

figure 6. New distribution system architecture with SIM.

IEEE power & energy magazine

47

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
ac Load UPS Battery

ac/dc

dc Load

figure 7. Traditional solution.

SIM

ac Load dc Load

Battery

figure 8. Solution based on SIM technology.

We live in a very critical and exciting time in the evolution of the electric power industry. 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. A smart grid will help the world manage demand growth, conserve energy, maximize asset utilization, improve grid security 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 technologies working together. Controllable technologies for supply, demand, power ow, and storage provide the means to implement decisions made by smart control algorithms 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 technologies as well as smart grid applications.

1) maximum utilization of the distributed generation to reduce the real power ow on the grid and 2) provision of reactive power where it is consumed to reduce the reactive power ow on the grid. The power electronicsenabled voltage regulation capability of SIMs will ensure a high-quality power supply at every load connection point by maintaining optimized voltage levels and compensating for voltage dips, swells, and ickers. The fast fault current detection and limiting capability will reduce the fault-breaking needs of circuit breakers on the feeder. The energy storage will provide a short- to medium-term power supply buffer so that customer service will not be interrupted in the event of short-term disruption on the distribution or transmission grid. This will relax the design requirements on the transmission grid.

For Further Reading

J. Berst, Why the smart grid industry cant talk the talk, Smart Grid News, Mar. 5, 2009. U.S. House of Representatives, (H.R. 6), Energy independence and security act of 2007, 2007. US Department of Energy (2008), The smart grid: An introduction, [Online]. Available: http://www.oe.energy.gov/SmartGridIntroduction.htm. European Commission, European smart grid technology platform, Luxembourg, 2006. Joint US-China Cooperation on Clean Energy (JUCCCE), Smart grid-future grid?A basic information report on smart grid, Dec. 18, 2007. National Institute of Standards and Technology (Sept. 2009), NIST framework and roadmap for smart grid interoperability standards, Release 1.0 (Draft), [Online]. Available: http://www.nist.gov/public_affairs/releases/smartgrid_interoperability.pdf Edison Electric Institute, Understanding smart grid. From definition to deployment, Washington, D.C., Mar. 2009.

Data Center Application

SIM technology could greatly simplify the supplying of power to data centers, a growing market segment, as well as improve reliability and reduce energy losses. Figures 7 and 8 illustrate this solution.

Incremental Transition Path

SIM technology is migration-friendly. It can be deployed incrementally and is compatible with the existing distribution system. This is a desirable characteristic, for it allows todays system to be transformed gradually into tomorrows intelligent system, the smart grid, by changing out traditional transformers one at a time.
48
IEEE power & energy magazine

Biographies
Enrique Santacana is president and CEO of ABB Inc. Gary Rackliffe is vice president of smart grids at ABB Inc. Le Tang is a vice president and head of the U.S. Corporate Research Center at ABB Inc. Xiaoming Feng is executive R&D consulting engineer at ABB Inc.s U.S. Corporate Research Center. p&e
march/april 2010