Warren Powell ’77, a professor of operations research and financial engineering, has developed a model to determine when to use renewable energy and batteries, alone or in combination.
Sameer A. Khan
Devising cost-effective ways to use renewable sources of energy
Warren Powell ’77, a professor of operations research and financial engineering, has developed a model to determine when to use renewable energy and batteries, alone or in combination.
Warren Powell ’77, a professor of operations research and financial engineering, has developed a model to determine when to use renewable energy and batteries, alone or in combination.
Sameer A. Khan

The Obama administration wants to decrease American reliance on fossil fuels by doubling the amount of wind and solar electricity we produce by 2025. That goal has energy producers studying when it’s practical to produce wind and solar energy, known as renewables, and when it’s not. Warren Powell ’77, a professor of operations research and financial engineering, is studying the best ways to integrate renewables and battery storage into the current energy grid.

Powell’s work addresses a challenge for the energy industry: relying on renewable sources that are highly variable and difficult to estimate. While the sun rises and falls regularly, predicting where and when cloud cover will appear is difficult. Wind energy is equally challenging. “Wind is complicated to forecast,” says Powell. “We don’t know what the atmosphere is doing every square meter at every second.”

Storing energy is one solution, but maintaining large amounts of energy in a battery for more than an hour or two is expensive. Furthermore, batteries store energy in direct current, known as DC, and power moves along the grid in alternating current, known as AC. The inverters currently used to switch between DC and AC are only 85 percent efficient. “You lose a lot,” Powell says.

Powell’s team has developed a model to determine when to use renewable energy and batteries, alone or in combination, given particular conditions, such as the amount of power available from elsewhere, the level of demand, and price fluctuations. Powell’s conclusions were based on a simulation that used the electricity grid serving 61 million people in the mid-Atlantic region.

The simulation suggested that, when price is considered, batteries are not cost-effective for long-term storage. Maintaining a store of renewable energy in large batteries and moving it for later use costs more than firing up fossil fuel-based turbines when energy is needed.

But battery storage is useful in one scenario, Powell says: When renewable-energy sources experience unexpected changes in wind strength or cloud cover, the fossil-fuel sources cannot react quickly enough to supply energy. In those cases, a battery can provide backup energy until other sources go online.

Ultimately, Powell’s model suggests, “storage is not the panacea people once saw,” he says. “Storing energy during the day to use at night is extremely expensive, and likely to stay that way for the foreseeable future.” Still, Powell’s work could lead to storing renewables in the right places at the right times.