ADVERTISEMENTS
Learn about diversity at ASEE
ASEE would like to acknowledge the generous support of our premier corporate partners.

 FEATURES

BY THOMAS K. GROSE
FEATURE: THE CYBER GRID

COVER STORY

THE CYBER GRID

Information technology will give America's electrical system a needed charge – but it won't happen overnight.


Information technology, which brought stronger, faster computers and the Internet to the masses, is set to overhaul America’s aging and rickety electric grid. Digital technologies are making it increasingly possible for utilities to balance power generation and load demands more efficiently “and do it in a quick and intelligent manner,” says Mariesa Crow, director of the Energy Research and Development Center at the Missouri University of Science and Technology (MS&T). In other words, America’s power grid is about to be dragged into the 21st century. It’s about to become a Smart Grid.

And it’s about time.

The grid, which nationally comprises three regional grids, is often called the world’s biggest man-made machine. And in many ways, it is an amazing creation, a testament to the engineers who cobbled it together, bit by bit. But that piecemeal legacy is also its undoing. The grid was not designed to any master plan, and it shows. It’s a hodgepodge that includes 16,000 power plants, around 3,300 utilities, and 300,000 miles of power lines. It is overly reliant on polluting fossil fuels – 40 percent of U.S. carbon dioxide emissions come from power plants. It’s inefficient, wasteful, and prone to failures – some catastrophic. An August 2003 blackout cut power to 45 million people in the northeastern United States and another 10 million in Ontario, Canada, all because improperly trimmed trees touched a power line in suburban Cleveland. Indeed, blackouts cost the U.S. economy $150 billion a year.

“It’s an antiquated system, and it’s fragile,” explains Mike Konrad, a chief architect at Carnegie Mellon University’s Software Engineering Institute, which oversees the Smart Grid Maturity Model. “It’s also not well-instrumented. It’s critical infrastructure, but it’s largely passive.”

That’s a scary description of a system we can’t live without. Electricity is essential to modern life. Our medical, communications, manufacturing, and transportation systems depend on it. We all expect it to be there when we need it – whenever we flick a light switch or turn on a computer. And our thirst for power isn’t slacking. Power consumption in North America is expected to jump from around 7 terawatts now to 15 to 20 TW by 2050 (a terawatt is a trillion watts). And building more coal-fired plants is environmentally untenable. That’s why the efficiencies a Smart Grid offers are so alluring.

Billions For New Projects

The Obama administration has made development of the Smart Grid a top domestic policy priority, albeit one reliant upon technologies that still need more testing or research. The $787 billion stimulus package Congress approved in February included $11 billion for Smart Grid projects and to train or retrain workers. The Department of Energy (DOE) has made available to utilities around $4 billion in matching grants to set up pilot Smart Grid projects. The Edison Electric Institute, which represents America’s shareholder-owned utilities, says nearly all of its more than 200 members have applied for the grants. Several projects are either already underway or about to start. Florida Power & Light, for example, is using stimulus funds to help finance a $200 million Smart Grid project in Miami. And Xcel Energy’s $100 million “Smart Grid city” project is bringing the technology to 50,000 homes in Boulder, Colo. The Federal Energy Regulatory Commission recently adopted policies to accelerate Smart Grid development, including the standardization of technologies to ensure it’s inter-operational.

If everybody who has [an electric-powered car] comes home around the same time and plugs in, it will crash the grid.- Steinar Dale, director of the Center for Advanced Power Systems at Florida State University

A Smart Grid will use digital information technology – the Internet, sensors, controls, and wireless devices – to make it more efficient, reliable, and better able to accommodate renewable energy sources, particularly solar and wind. Renewables are intermittent producers of power – winds die down, the sun sets every evening – and that can bedevil an electric network that needs to be ever ready to meet fluctuations in demand with consistent, quality power. Building a Smart Grid is all about using IT to control load so that it better matches generation. That makes it easier to be more reliant upon renewables. Moreover, if demand peaks can be flattened, there is less need to build extra power plants to meet the few occasions a year when peak loads occur. “You either control load or have standby [fossil fuel] generation, which is not helpful to the environment,” says Steinar Dale, director of the Center for Advanced Power Systems at Florida State University. DOE says that just a 5 percent increase in grid efficiency would have the same effect as eliminating the fuel consumption and emissions of 53 million cars.

Communication and Security

Controlling load to achieve efficiencies also requires two-way communications – the ability for digital “smart meters” and “grid-friendly” appliances to communicate with utilities and vice versa. Then, when demand soars and the cost of electricity rises, non-vital appliances like dishwashers and dryers can be told to shut down until nighttime, when demand ebbs and electricity is much cheaper. Hot-water heaters can be notched down a few degrees and air conditioner temperatures set a tad higher. “It is all IT,” Dale says, “but it has to be embedded into the microchips of the appliances.” It also has to be robotic, so that changes in settings are done automatically and don’t require the customer to do anything, keeping it user-friendly. “But we will also have to deal with customer acceptance, to allow utilities to do this,” Dale says. For most customers, the promise of much cheaper electric bills should be incentive enough. Ultimately, customers who install solar panels or mini-wind turbines to reduce their dependency on the grid should be able to sell electricity back to it.

“The killer app is being able to have a controllable power and information interface. That is what we are working toward,” Crow says. It’ll also be a massive management problem, since power generation will be decentralized. That means relying on the Internet. “It will be a very complex communications system,” says Robert Lasseter, a professor emeritus of electrical and computer engineering at the University of Wisconsin – Madison. “But we can put the Internet through power lines – all of this is doable.” An IT-infused grid could also be self-healing, able remotely to detect and react to problems before they become catastrophic, with very little, if any, human input.

Of course, the grid is not only susceptible to equipment failures and the whims of Mother Nature; it's a potential terrorist target. The Wall Street Journal reported in March 2009 that Chinese and Russian spies had already hacked into grid communications systems. Yet, we're aiming for an Internet-based, decentralized grid that accommodates distributed energy sources. That will give the computer-savvy bad guys hundreds of thousands of entry and observation points instead of just a few. "It's significantly more difficult to protect," says Bruce McMillin, an information security expert at MS&T, especially since that would require confidentiality and obfuscation, concepts historically alien to the power industry. When power engineers talk about security, McMillin says, they're thinking of system stability, not information security. He's not entirely sure a Smart Grid can be fully secure, cyberly speaking. "It's going to be hard," he says. Good, immediate detection will be key to the effort. Arun Somani, an electrical and computer engineer at Iowa State University, says his team is making progress on a grid-monitoring network that would use wireless sensors to constantly keep tabs on the system's health, instantly detecting failures ranging from fallen trees to sabotage. The government clearly takes the issue seriously: Utilities that receive those matching DOE grant funds for pilot programs must show they're working on cybersecurity.

Northeast Blackout of 2003: NOAA satellite imagery one day before and the night of the blackout

One security advantage of a Smart Grid is that it may also incorporate a network of microgrids, mostly powered by renewable energy sources. If a problem occurs on the main grid, affected microgrids can disconnect themselves automatically - becoming islands of power - then reconnect later, once the fault is fixed. If a microgrid can't generate enough power to meet local needs, it can shed load by temporarily cutting power to nonessential appliances, like washing machines, and reducing power to smaller appliances, like fans and toasters. But power to essential users, say, hospitals and critical computers, would remain uninterrupted. Ergo, if there are thousands of microgrids in a region, a terrorist attack on a transmission line won't do as much damage as an attack on a centralized system. "It makes for a less attractive target," explains Lasseter, a leading microgrid expert.

Hot Field: Battery Technology

Beyond security, storage is a major concern, too – perhaps the biggest. Solar panels and wind turbines are capable of producing an excess of electricity, so it’s essential to find ways to store massive amounts of that excess power for use later on. Unfortunately, there is no good, cheap way to store huge amounts of electricity – battery technology has hardly advanced in 40 years. “A breakthrough is needed,” says Iowa State’s Somani, “but it is not at all predictable when it might happen.”

Well, it might happen sooner than expected. Battery technology has become a hot area, with millions of investment dollars pouring in. General Electric has plans for a plant in New York to build huge batteries, while IBM says it’s begun development of a next-generation battery. GE is initially investing $100 million in the upstate plant. Although much research these days focuses on lithium ion batteries, GE is going to produce an updated version of the sodium sulfur battery, which is popular in Japan for storing grid electricity. The appeal of this battery is its simplicity: Its key components are liquid sodium and salt. Meanwhile, states like Michigan, hard-hit by the recession, are hoping to create new jobs by wooing battery companies with tax breaks and other incentives. Universities are getting in on the act, too, with many, like Flint, Mich.’s, Kettering University creating new battery technology courses.

General layout of electricity networks. Voltages and depictions of electrical lines are typical for Germany and other European systems. - J JMesserly

Donald Sadoway, a professor of materials chemistry at the Massachusetts Institute of Technology (MIT), has developed an all-liquid battery – its electrodes are molten metal – that operates at electrical currents 10 times higher than any other known battery. The materials required are inexpensive, and Sadoway believes that large cells of his liquid-battery packs, wired together, will have the capacity to essentially “store the grid.” Researchers are also looking at other types of storage technologies, including compressed air, ultracapacitors, thermal energy storage, and fuel cells. Some of these technologies aren’t new, such as compressed air, which uses excess power to compress and shunt air into underground storage. Later, the air is released to power turbines. A more radical approach is the ultracapacitor, an update on the capacitors used in car engines to give them quick boosts of power. Capacitors can store and deliver energy rapidly but haven’t been used to hold large amounts of energy before. An invention by University of Texas - Austin researchers led by Professor Rod Ruoff, who heads the school’s mechanical engineering department, uses a new nanotech material they devised from carbon called graphene. Because it’s ultra-thin, many layers of it can be stacked together to store huge amounts of energy. The technique has led to the creation of a spinoff company, Graphene Energy.

It will be a very complex communications system, but we can put the internet through power lines - all of this is doable- Robert Lasseter, professor emeritus at the University of Wisconsin- MadisonBut many experts, including those at DOE, believe that the real storage solution will be so-called vehicle-to-grid (V2G) technology. Eventually, when there are millions of plug-in electric hybrid cars on the road, all those car batteries could be used to store excess electricity. When a car is parked – which is often – it’s plugged into a socket, and if its battery has more charge than it needs, it can sell electrons back to the grid. A typical plug-in car can produce more than 10 kilowatts of electricity, enough to power around 10 houses. Research teams at the Universities of Michigan and Delaware are focusing on V2G technology. The Ford Motor Co. recently announced that its future plug-in hybrids will be able to “talk” to a Smart Grid. Ford is investing $14 billion in the project and has worked with 10 different utilities, DOE, and two research institutes to make the cars grid-compatible. It’s a big step forward. Customers will save money because they’ll be able to charge their cars when electricity rates are lowest, and utilities will be better able to manage load. Still, none of these first-generation Ford plug-in hybrids will be able to send electrons back to the grid, the ultimate goal of V2G technology. That capacity is still a few years away from becoming reality. And the main hurdle is coming up with software to accommodate millions of folks plugging in at any given time, says Florida State’s Dale. “If everybody who has a car like that comes home around the same time and plugs in, it will crash the grid.” The intelligence, he says, will have to be in the car itself.

Advances Will Be Piecemeal

Storage can also be done on a smaller scale, at the microgrid level. MIT researcher Daniel Nocera has come up with a cheap, clean, and efficient way to split water into hydrogen and oxygen using electricity from solar panels or mini-turbines. The gases can later be reformed in a fuel cell to produce electricity. That’s right, electrolysis. Current electrolyzers use platinum as a catalyst, “which is really bad, expensive, and inefficient,” says Nocera, a professor of energy. His version uses cobalt metal, phosphate, and electrodes as a catalyst. Because it is a technology that works best with relatively small amounts of electricity, “this is definitely compatible” with a microgrid concept, Nocera says.

So, how long will it take for a nation-spanning Smart Grid to be fully realized? Probably 10 to 15 years. And, of course, it won’t happen all at once but instead, section by section, utility by utility. As Ed Legge, spokesman for the Edison Electric Institute, says, “It will be absolutely piecemeal; there’s no other way to do it. This is a system that almost literally serves every person on the continent. It is an enormous logistical undertaking.”

Moreover, bringing smarts and digital technology to America’s aging and creaky national grid system will cost an estimated $65 billion. Customers will pay for most of this huge upgrade. But if it’s done properly so that utilities have more “robust and granular” control of the systems and demand peaks can be shaved down, it should eventually save customers money, says Legge. Engineers at Virginia Tech say that a fully operational Smart Grid could cut capacity by 13 percent. “The idea is that ultimately, it pays for itself,” Legge says.

Given the urgent need to make the grid system more robust, greatly reduce carbon emissions, and cut U.S. dependence on foreign oil, the Smart Grid certainly looks like a sound investment. “Necessity is the mother of invention,” MS&T’s Crow says, “so I believe this will happen. Twenty years ago, no one could have predicted what the Internet would do for us. So I can’t say what [our electricity system] will look like in 20 years. But I know it will be very different.” And almost certainly a heck of a lot smarter.


Thomas K. Grose is Prism's chief correspondent, based in the United Kingdom.

 

TOPˆ

 


ASEE
© Copyright 2009
American Society for Engineering Education
1818 N Street, N.W., Suite 600
Washington, DC 20036-2479
Web: www.asee.org
Telephone: (202) 331-3500