Skip to main content

Cloudy With A Chance Of Electrons: This Scientist Can Forecast Renewable Electricity

Todd Alhart
February 29, 2016
A group of physicists that included a Nobel laureate and Kurt Vonnegut Jr.’s brother spent the 1940s working in GE labs to figure out how to control the weather. After a promising start – they created snowfall over Schenectady, N.Y. – the project eventually fizzled.
Seven decades later, GE electrical engineer Naresh Acharya is attempting something similar. His system could one day help utilities predict the amount electricity generated by inherently unpredictable renewable resources like wind and sun, and send as much of it as possible over the grid to consumers.

Working at GE Global Research in Niskayuna, N.Y., not too far from Bernard Vonnegut’s lab, Acharya is using software to predict and balance the electricity generated by renewables and the amount of power used by consumers. The research is partially funded by the U.S. government’s Advanced Research Project Agency – Energy (ARPA-E).“The goal of ARPA-E’s program is enable a grid that can reliably manage a power mix where nearly half or more is supplied by renewables,” he says.

The software must aggregate and control thousands of customer loads in real time and match them with production projections. “We want to create a scenario that never puts a damper on the grid or the clean power that’s delivered to homes,” he says.

GRC Naresh Weatherman-17 “The goal of ARPA-E’s program is enable a grid that can reliably manage a power mix where nearly half or more is supplied by renewables,” says GE's Acharya (above).

It’s a sky-high challenge. Today, most of the power that U.S. utilities provide comes from traditional centralized power plants that run on fossil fuels. These plants often provide some reserve capacity that can be readily controlled by a grid operator. Since the load changes every second, this reserve capacity is very handy in managing these fluctuations. But as more fickle renewable power is brought online, providing the reserve capacity for the entire grid will become more challenging. The sun doesn’t always shine, and the wind doesn’t always blow.

The stakes are also quickly rising. In 2015, more than 65 percent of America’s new generating capacity came from solar and wind energy. The day when renewables could supply half of all electricity may be closer than you think.

Acharya and ARPA-E aren’t working on the challenge alone. Their other partners include GE Energy Consulting, Lawrence Berkeley National Laboratory, Enbala Power Networks, Consolidated Edison, Inc., Southern California Edison, Sacramento Municipal Utility District and California Independent System Operator. He will present his results at ARPA-E’s annual Energy Innovation Conference this week in Washington, D.C.

Acharya says that in recent years, renewable power was the global leader in bringing new generation online, adding more than 100 gigawatts of wind and solar in 2014 alone. This amount could power the entire state of Texas. Moreover, countries like Denmark are already generating 40 percent of their electricity from wind.

“We need to get visibility to bring all this energy into our home,” Acharya says. “We can’t control the weather, but our software will help us control the electricity it generates. I can live with that.”

NB10460, Cloud seeding, Langmuir, Bernard Vonnegut, Vincent Schaefer, 1947 Nobel laureate Irving Langmuir (left), Bernard Vonnegut (center) and Vincent Schaefer are seeding snow clouds in GE labs in 1947. Their effort to control weather fizzled out. Now another GE scientist is using software to predict and control electricity generated by the wind and the sun. Image credit: Museum of Innovation and Science Schenectady