Why we may need to think small — not big — to green the energy grid

September 16, 2020

A UM-Dearborn graduate student is working to fulfill the promise of electric “microgrids.”

A collage graphic representing energy grids with lots of renewable energy from wind turbines and solar panels.
A collage graphic representing energy grids with lots of renewable energy from wind turbines and solar panels.

The prospect of bringing more renewable energy onto the grid holds a lot of promise for dealing with the challenges of climate change. But an energy system that relies more heavily on solar and wind power also presents some big technological issues — especially once renewables reach a critical mass. The major challenge stems from their fundamental intermittence: If there are unexpected periods where the sun isn’t shining or the wind isn’t blowing, it could cause major disruptions to energy supplies — meaning the always-on reliability of the grid could be at risk.

Associate Professor of Electrical and Computer Engineering (ECE) Wencong Su and ECE doctoral student Fangyuan Chang are among the researchers looking for solutions to big-picture problems like this, one of which is decidedly small. Their latest project focuses on energy “microgrids,” which are pretty much what they sound like. In contrast to the single massive electric grid we use today, where power is produced at big power plants and then sent all over the country, microgrids produce smaller amounts of power that is typically consumed more locally. It’s not an idea that’s complete science fiction. In fact, microgrids are being used today to power everything from data storage centers in Japan to factories in Michigan to several small American cities. 

Chang says this surging interest in microgrids is coming from a variety of places. The U.S. military, for example, is interested in microgrids because they offer the promise of energy security and independence. Big manufacturers like microgrids because they allow them to produce more of their own energy — and to do it from renewable sources, which helps corporations with sustainability goals meet their targets. And new interest in microgrids that use DC energy, in contrast to the standard AC energy that electrifies the grid today, is being driven by technological change: Things like solar panels, electric vehicles, electric storage systems, cell phones and computer CPUs all operate on DC not AC energy. Because of this, a microgrid that produced and used DC energy directly, without any need for conversion, would be inherently more efficient.

Fangyuan Chang and Associate Professor Wencong Su
Fangyuan Chang and Associate Professor Wencong Su

Interestingly, Chang says this new excitement around smaller DC-based grids, which is the focus of her own work, is reviving a more-than-century-old debate about which type of electric current is best. “Starting in the late 1880s, Thomas Edison and Nikola Tesla were involved in a battle known as the ‘War of the Currents,’” Chang explains. “Tesla’s alternating current, or AC, eventually became dominant in the electric power industry for the past 100 years, so there was little interest in research in DC [direct current] microgrids. But now we’re seeing a bit of a renaissance.” 

Realizing the full potential of that renaissance, however, will depend on researchers like Chang solving some existing issues with DC-based grids. Stability, she says, is one of the biggest challenges. Just as a sugary drink can cause a person’s blood sugar to spike, surges (or drops) in local energy production and consumption within DC microgrids can lead to instability and poor quality electricity. To smooth out those highs and lows and keep the voltage in an optimal range takes sophisticated control systems that run on complicated math. And it was while working on that part of the problem that Chang made her biggest discovery to date. 

“Originally, my plan was to come up with an application based on an existing stability theory for nonlinear circuits, but when I built a simulation, I found something weird,” she explains. “According to the previous theory, my simulation should have been stable, but it wasn’t.” At that point, Chang dug deeper — ultimately discovering that the 60-year-old mathematical concepts underlying a lot of DC energy research were flawed and incomplete. At that point, she began working on her own new theories, which are the basis of research she and Su have recently published in a prestigious IEEE journal

Discovering flaws in a longstanding theory would be a major development, and Chang is already pushing ahead with research that could provide evidence that her new approaches really do work better. She and Su have developed new control methods for DC microgrids and stability criteria based on Chang’s new math, and so far the results have proved very promising in simulation. Now, they’re preparing to take a big next step: Testing them in real-world microgrid systems through a collaboration with colleagues at the University of Texas-Austin and funding from the National Science Foundation.

“It is very exciting for us because from our side, we can already verify our theory in simulation, so I think we are onto something really different,” Changs says. “But you have to know it will be effective in a real-world environment — where anything can still happen.”