The promise of renewable energy sources such as solar and wind is that they’re increasingly cheaper, cleaner and easier to harvest. But they’re also fickle, so that a period of cloudy skies or calm winds can greatly reduce the amount of energy they can generate.
IBM says it’s solved this problem, up to a point. No technology company can ensure sunny, breezy days, but IBM says it and the National Renewable Energy Laboratory (NREL) have developed a program that tailors weather forecasts to give renewable utilities information about what to expect from the weather and how it will affect power output.
What’s more, IBM and NREL will provide these forecasts for free to utilities throughout North America.
Related: MLPs Under Threat From The IRS
In announcing the development on July 17, IBM Research said the technology involves cognitive computing technologies, including machine learning, to make such weather forecasts up to 30 percent more accurate than conventional methods.
“By continuously training itself using historical records from thousands of weather stations and real-time measurements, IBM’s system combines predictions from a number of weather models with geographic information and other data to produce the most accurate forecasts – from minutes to weeks ahead,” said Dr. Siyuan Lu, an IBM researcher.
This increased precision could help utilities avoid generating more power than they need – hundreds of megawatts each year, in some cases – and reduce their need to build special conventional plants to supply extra power during periods of peak demand. This could lower toxic emissions and save the companies millions of dollars, savings that they could then pass on to their customers.
Related: Obama’s Goals For Middle East Hinge On Putin
The NREL says a study of such forecasting that was only 25 percent more accurate than conventional methods could save as much as $46.5 million a year for one utility, which it didn’t name, that serves the New England region of the United States.
“What we are doing is combining multiple models together into one ‘supermodel,’” said Hendrik Hamann, a research manager at IBM. Such a program can base forecasts in part on the location of the forecast area, its weather history and current conditions, he said.
This is particularly important for solar energy generators, which often face sudden drops in energy production. For example, if an unexpected cloud obscures the sun, it could quickly drain a solar array of as much as 70 percent of its ability to generate electricity.
Related: This Is Why Californians Pay More For Their Gasoline?
The program also would be important for customers with their own solar installations that are parts of distributed-generation grids. Using the IBM forecasts, they can share power more efficiently.
Solar power was the fastest-growing form of electricity generation in the United States in 2014, according to the Solar Energy Industries Association, but further expansion would be enhanced by more accurate weather forecasts.
“By improving the accuracy of forecasting, utilities can operate more efficiently and profitably,” said Bri-Mathias Hodge, director of the NREL’s Transmission and Grid Integration Group. “That can increase the use of renewable energy sources as a more accepted energy generation option.”
By Andy Tully of Oilprice.com
More Top Reads From Oilprice.com:
Andy Tully is a veteran news reporter who is now the news editor for Oilprice.com