You are here

Beyond "Partly Sunny": A Better Solar Forecast

December 7, 2012 - 10:00am

Addthis

The Energy Department is investing in better solar forecasting techniques to improve the reliability and stability of solar power plants during periods of cloud coverage. | Photo by Dennis Schroeder/NREL.

The Energy Department is investing in better solar forecasting techniques to improve the reliability and stability of solar power plants during periods of cloud coverage. | Photo by Dennis Schroeder/NREL.

A passing cloud can make a big difference in a solar power plant’s output.

That’s why the Energy Department is funding research to help utility companies and grid operators better predict and plan for changes in the solar resource -- which means more efficient and reliable energy for consumers.

As part of the SunShot Initiative, the Department awarded $8 million to two projects that will create enhanced tools to forecast when, where and how much solar power U.S. energy plants will produce.

Because a solar power plant's electricity output depends directly on the amount of sunlight that hits the solar array, changes in the weather can cause dips in power production. Improved forecasting technologies will help utilities and power system operators better predict when clouds and other weather-related factors will reduce the intensity of incoming sunlight.

In turn, this information will allow utilities and operators to more accurately anticipate changes in solar power production and take actions to ensure the stability of the national power grid. Because accurate forecasting can also help utilities and grid operators boost the reliability of their systems, it can ultimately reduce the cost of integrating solar into the grid.

Two teams comprised of industry, national laboratory and university partners will explore solar forecasting solutions under this program:

  • The University Corporation for Atmospheric Research and its partners will advance methods for measuring solar radiation, observing clouds and predicting impacts through nowcasting. The research team will also develop approaches to quantify and track aerosols, haze and contrails that affect cloud formation and develop short-term prediction techniques of cloud properties based on observations.
  • The IBM Thomas J. Watson Research Center and its partners will integrate big data processing and cloud modeling into a universal platform that combines different prediction models and uses state-of-the-art machine learning technologies to drastically improve the accuracy of predictions. Similar to the recently demonstrated IBM Watson computer system, the proposed Watt-sun technology will leverage deep machine learning and self-adjusting voting algorithms to decide between various forecasting models and expert systems.

These SunShot projects constitute a public-private partnership between the Energy Department, the National Oceanic and Atmospheric Association (NOAA) and the awardees to improve the accuracy of solar forecasts.

Find out more about the Department’s SunShot Initiative and learn about how we’re working to reduce the cost of solar energy by 75 percent by the end of the decade, making it cost competitive with other forms of American energy. 

Addthis