Increasing Power's Reliability During Natural Disasters

Assistant Professor Alexis Kwasinski is studying how different power infrastructures react to different natural disasters. This photo, taken by Kwasinski while assesssing damage in Japan, shows a downed power line. Nearby, however, a church gets power from solar panels.

Assistant Professor Alexis Kwasinski is studying how different power infrastructures react to natural disasters. This photo, taken by Kwasinski while assesssing damage in Japan, shows a downed power line. In the background, however, a church gets power from solar panels. View more of Kwasinski's photos.

Research by a professor at the Cockrell School of Engineering could lead to more reliable power and communications systems during natural disasters, an advancement that would enable emergency responders and families to locate survivors faster and that could prevent some of the fatalities and injuries that occur.

Research in this area has garnered the attention of telecommunications giants, like AT&T and Sprint, who want to prevent disruptions to mobile and phone communications, as well as governmental agencies, which seek to ensure that critical facilities – like hospitals and military bases – don't lose power during a natural disaster or attack. The need for more reliable power has been underscored this year by earthquakes in Japan and New Zealand and severe storms that ripped through the Midwestern United States this summer, knocking out power and communications systems along the way.

Alexis Kwasinski.

Alexis Kwasinski. Click to enlarge this image.

In an effort to understand and prevent such failures, Department of Electrical and Computer Engineering Assistant Professor Alexis Kwasinski is studying how different power infrastructures react to different natural disasters. The research has taken him to U.S. coasts, where he assessed what worked and what failed in the power chain following hurricanes Katrina and Ike, and to Chile, New Zealand and Japan, where he is currently collecting data.

Kwasinski's research focuses on power and communications infrastructure with special attention dedicated to microgrids: small, decentralized power grids that generate and store their own electricity through a mixture of alternative power sources, typically solar, microturbines, fuel cells and natural gas. Research on these self-supporting grids is gaining traction around the world because they are considered by some one of the building blocks of smart grids and, most of all, because of their distinguishing feature: microgrids can separate and isolate themselves from the main utility grid when there is a disturbance with little or no disruption to their own power supply. Since the utility grid is a secondary source, the microgrid is protected from its surges and failure and can automatically reconnect to the utility grid once the disturbance is resolved.

"A single downed line in the utility grid can cause significant power and communications outages following a natural disaster," Kwasinski said. "That's why a microgrid, and its ability to disconnect itself from the main grid while still generating its own power, is so important to hospitals, military bases, telecommunications companies and any other facility that can't afford to lose power."

Power outages after Hurricane Katrina demonstrated the need to improve power and communications reliability in the wake of natural disasters. Kwasinski said developing and improving the infrastructure of microgrids is just one way to do so.

He is currently in the midst of a five-year research project funded by the National Science Foundation and for some of the damage assessments also supported in part by the American Society of Civil Engineers. The research teams Kwasinski with civil engineers, who are analyzing the performance of other critical infrastructures, such as roads, and also studying how to rebuild and remove debris.

The recovery and rebuilding process in Japan is daunting, Kwasinski said, but the country offers the best and worst examples of what can happen to the power system when a natural disaster occurs and when a microgrid is in place.

For instance, after the earthquake the Fukushima Dai-ichi nuclear plant lost power from the main utility grid and emergency power generators were left inoperable following the tsunami. The loss of power prevented the reactor's cooling systems from working and led to a nuclear radiation leak. Some 70 miles away, however, a microgrid in Sendai, Japan remained unaffected and supplied power to a hospital and part of an university campus connected to it.

"A severe earthquake or hurricane can damage any infrastructure but functional diversity with different energy sources limits the vulnerability of a power and communications system," he said.

Smarter design of microgrids also helps reduce vulnerability.

Kwasinski traveled to Japan this week for his second trip since the March 11 earthquake. Based on an analysis of the data he collects there, he aims to discover how different types of infrastructure interacting with a microgrid react during natural disasters.

“In a critical event like a hurricane or earthquake, you see how these interdependencies [between power supplies] work more clearly than you would in normal circumstances, so it's kind of like an extreme test," Kwasinski said.

The idea, he said, is to determine the best way to design and build a microgrid based on the types of natural disasters it is likely to face.

For example, he said, if a microgrid is located in an area prone to hurricanes, natural gas can be used as a main power supply to the small grid because the pipes run underground and are not easily damaged by hurricanes. But for microgrids in areas prone to earthquakes, natural gas is not as viable of an option because the gas is usually interrupted by fire or leaks caused by the disaster.

Kwasinski’s research is reactive for the moment, as he travels to disaster areas to assess damage. But what he gleans from these visits could lead to power supplies that are less vulnerable, rely on multiple energy sources and are proactively designed to withstand natural disasters.

He illustrates this point with a recent example from Japan, where emergency text messages were sent immediately after the first seismic waves struck before the main earthquake.

“If the main earthquake takes out the telecommunications facility, then you cannot have this emergency notification service for the aftershock,” he said. “And that can be as dangerous as the main earthquake because buildings have been weakened. A microgrid helps reduce the opportunities for power failure.”