Working toward a smarter grid

Researchers collaborate to develop technologies for tomorrow’s energy infrastructure.

Nothing brings home the value and vulnerability of electricity like five days without power. That’s what students, faculty and staff at Lehigh — along with people in more than 100,000 homes and businesses in the Lehigh valley and millions across the northeastern United States — experienced after a late- October snowstorm in 2011.

Just weeks earlier, Hurricane Irene and the remnants of Tropical Storm Lee had caused other major outages across the eastern U.S.

Back-to-back bouts of unplanned urban camping prompted many Americans to ponder the infrastructure that generates and delivers our electricity. How does this grid work? What happens when it doesn’t? And how can it be made to function better?

A group of Lehigh faculty and students is tackling these questions, and the temporary blackouts renewed the team’s determination to help usher in a complex and dramatically improved future for the grid.

These researchers imagine a future in which a freak weather event will no longer paralyze an entire region of the country, leaving residents to light candles, turn on flashlights and wait for repair trucks to come. Instead, autonomous “micro-grids,” connected to the main power grid but able to operate independently if power is disrupted elsewhere, will continue generating electricity until primary distribution is restored.

Meanwhile, monitoring and communication systems throughout the grid will detect where electricity is needed, where it is available and where disturbances and faults exist, and then trigger a self-repair process that keeps the lights on.

Drawing on islands of power, and enabling elements of the grid to talk back and forth, are two parts of a larger vision for transforming the nation’s electrical landscape to create what has popularly come to be known as the smart grid.

“The idea is to overlay an information network like the Internet on top of the power grid that we’re familiar with,” says Rick Blum, the director of a new research cluster called Integrated Networks for Electricity (INE). “This will allow us to create an exciting new environment with great potential for giving electricity providers and consumers a level of control that’s never existed before.”

A smart grid will enable real-time communication between utilities and consumers, which will in turn allow energy users to know when peak power demand makes electricity relatively expensive — and motivate them to use it when it’s cheaper. Smart appliances plugged into the grid will detect energy price fluctuations and operate automatically at the most economical times.

The smart grid will also integrate energy from almost any source, better accommodating alternative and green technologies such as wind and solar power. If you have energy to spare from your rooftop photovoltaic array, backyard wind turbine — or even the batteries in your electric car — you will be able to sell it back to the grid. Electricity consumption would be evened out through the 24 hours of the day. As a result, utility companies could reduce their reliance on the expensive, higher-polluting back-up boilers that they must keep ready to meet spikes in demand. Power generation and distribution would become cheaper, cleaner, greener, more efficient and more reliable.

It’s not a pipe dream: A total of $11 billion in smart-grid investments was allocated in the 2009 American Recovery and Reinvestment Act. A massive upgrade of our electrical infrastructure is already overdue. Peak demand for power to feed energyhungry air conditioners, electronics and appliances has exceeded growth in transmission capacity by 25 percent a year since the early 1980s, according to the U.S. Department of Energy (DOE). A 2009 report by DOE’s Electricity Advisory Committee concludes that the nation’s energydelivery infrastructure “will be unable to ensure a reliable, cost-effective, secure and environmentally sustainable supply of electricity for the next two decades.”

Making the smart grid a reality, says Blum, the Robert W. Wieseman Professor in Electrical Engineering, will require development of what some call the “Enernet.” This massive yet decentralized new system will allow the easy, reliable and simultaneous flow of three essential commodities — electricity, information and money.

“The core technologies that are needed to develop the smart grid include information technology, microelectronics and communication networking,” says Blum. “These intersecting areas fit precisely with Lehigh’s strengths and give us significant opportunities to collaborate with industry.”

A team for tomorrow

The development of smart-grid technologies demands input and perspective from diverse fields, including computer science, mathematics, industrial and systems engineering, electrical engineering and economics.

To leverage the strengths of researchers in those areas, Lehigh established the Integrated networks for Electricity cluster in 2011. The new research cluster will work with Lehigh’s Energy Systems Engineering Institute (ESEI) and Energy Research Center, which study energy generation, distribution and consumption, and their environmental impact.

“The research cluster concept is to bring faculty from different disciplines together to collaborate and support each others’ projects, processes, research and programs,” says ESEI director Martha Dodge. “Multidisciplinary teams are expected to play a key role in developing advanced electricity systems of the future.”

To fill out its team, the InE cluster is adding four new faculty positions — in energy economics, applied mathematics, real-time engineering systems and power system electronics.

“This is a great example of where collaboration will bear fruit,” says Blum. “Lehigh has always been strong in bringing ideas together, and our faculty are in the habit of working across disciplinary borders on important problems.”

Students in Lehigh’s innovative 10-month professional master’s of engineering degree in energy systems engineering (ESE) work on projects mentored by faculty in the InE research cluster. Just as important, both students and faculty work on real projects supplied by industry.

“It’s a unique program,” says Dodge, a former senior administrator at PPL Electric Utilities. “One, it’s relatively short — most similar programs are two years. But it’s also taught by both academic and adjunct professors from industry. That dual emphasis produces engineers with both technical and economic perspectives who are prepared to work in a business environment.”

In another innovation, the ESEI has established a program called the Keystone Smart grid fellowship through a DOE grant. Its goal is to attract K-12 and community college teachers who want to improve their understanding of the modern power industry and prepare their students for smart-grid careers.

“The number of engineers in the energy industry is diminishing,” Dodge says. “Developing the workforce is critical.”

Envisioning the future of energy

One current smart-grid project trains ESE students in the concepts of renewable energy. With DOE funding, an initial group of students simulated an “energy island” capable of shielding a college campus (or residential neighborhood, office park, hospital or military installation) from the ravages of power outages. Known as integrated renewable energy parks, or IREPs, these microgrids also harness power from wind, solar and other renewable sources for use in a small geographic area.

To determine if the idea would be cost-effective, students first collected information about renewable technologies, including prices for elements such as solar panels. One student, a high school teacher with a degree in civil engineering, looked at setup costs such as installation. A student with a degree in mathematics and experience in economic analysis helped evaluate an IREP’s appeal to investors.

The end result was a software package that provides investors with analyses of different IREP types and configurations.

“The location of an IREP is very important,” says Liang Cheng, associate professor of computer science and engineering. “The students’ software made it possible to advise investors on the pricing of a microgrid, as well as how much power could be generated by solar and wind at a particular site.” A second student team later refined the software.

Wireless communications and networking are expected to carry much of the monitoring and control information that will pass through IREPs and other areas of the smart grid. Yet the wide range of energy sources and devices that communicate within the grid could crowd available bandwidth.

Part of Cheng’s research explores the use of what’s known as reconfigurable radio (RR) — wireless communication hardware and software that can adjust and adapt to a spectrum of uses. If a particular bandwidth is not available, RR can change a device to a frequency that is being underused. These concepts are covered in a course called Communication networking Technologies for the Smart grid that Cheng co-teaches with Shalinee Kishore, associate professor of electrical and computer engineering.

The practicality of IREPs and other smart-grid concepts will depend on energy storage or backup supplies. A microgrid’s renewable sources have to provide power when the sun isn’t shining or the wind isn’t blowing, which means energy must be stockpiled so that it can be drawn when needed from a reserve.

Ensuring backup supplies requires contracting reserves in advance from conventional sources such as gas, coal and nuclear, says Alberto Lamadrid, a new assistant professor of economics who has a bachelor’s degree in electrical engineering and a master’s in operations.

It also involves patterns of energy production and use. A person charging an electric car at 2 a.m., says Lamadrid, “is in effect buying energy during off-peak hours and deferring its use until a higher-demand time.”

Or consider wind power. It requires large-scale storage that is still relatively expensive because wind turbines often harvest the most energy in hours after midnight when consumption falls off. But consumers eventually may adopt newer, more flexible approaches to energy use that help balance electricity supply through the use of energy management controllers (EMCs) — devices that make decisions about energy use in response to demand and price.

Another potential energy source for the smart grid that Lehigh researchers are studying is the flow of waves in the deep-water ocean. Unlike tidal power systems, which rely on structures that are easily seen from shore, wave-energy systems have a much smaller visual profile. And unlike the sun and wind, ocean waves generate energy in a manner that is more steady and predictable and less time-dependent.

“Several wave energy devices have been deployed worldwide, but mostly in isolation,” says Larry Snyder, associate professor of industrial and systems engineering, who is studying wave energy with Kishore, Blum and other faculty.

“The real potential of wave power will be realized only when multiple devices are deployed together in a ‘wave farm.’ There are many questions about how best to design and operate these farms, and those questions are the focus of our research.”

The group is developing quantitative analysis models to try to predict the buoys’ behavior and determine the optimal configuration of the buoys for generating electricity. One challenge is to “stagger” the up-and-down movement of the buoys to produce the steadiest output of power. Another is to determine the most effective way to connect the buoys.

“We are tapping into the relatively large knowledge base regarding wave motion,” says Snyder. “It’s exciting because there’s a lot of interest in this area but also a lot of work to be done.”

Safeguarding the grid

Research into the smart grid, says Blum, takes place under a shadow of concern about the security of the computer networks that orchestrate the communications that enable modern society to function.

“A hacker who intercepts data as it’s transmitted and changes it could cause significant problems,” says Blum. Similar threats have been documented — or at least reported — in other energy-related systems, including viruses found in grid computers in the U.S. and viruses that have reportedly done extensive damage to systems in Iran’s nuclear program. Blum uses multiple sensors to measure signal processing and detect data security breaches and failures in smart-grid systems. While sensors measure variables such as current and power, and the speed at which a generator turns, mathematical models describe and predict their behavior throughout the interconnected network.

“Measurements can’t jump to a strange value,” Blum says. “If you see inconsistencies between what the mathematical models predict and what the sensors detect, you have to determine if something has happened or if there’s just an honest fault. Did something break — or did someone break in?”

Cheng’s research approaches the security question from a different angle — protecting the boundaries of a network by encrypting packets of data.

Blum uses multiple sensors and signal processing to measure and detect data security breaches and failures in smart grid systems. The key, they say, is to combine academic research with industry collaboration and move from theory to implementation.


Originally published at www.lehigh.edu.