HyRef Technology Revolutionizes Renewable Energy Forecasting

Sign up for daily news updates from CleanTechnica on email. Or follow us on Google News!

IBM has long been known for building some of the world’s most powerful supercomputers, but what happens when it applies advanced modeling to solving the intermittency of renewable energy?

The answer, it turns out, is “Hybrid Renewable Energy Forecasting” (HyRef). This new technology, already online in China, is able to produce accurate local weather and renewable energy forecasts as far as one month in advance, down to 15-minute increments.

The HyRef technology combines advanced power and weather computer modeling, sophisticated cloud imaging, sky-facing cameras, and on-site sensors to accurately predict solar power and wind energy output and increase the amount of renewable electricity flowing onto grids across the world.

HyRef forecast system
HyRef forecast system graphic via IBM

Crowded Field In Renewable Energy Forecasting Race

HyRef joins an increasingly crowded field of innovative technologies seeking to accurately predict the output of renewable energy resources. The National Center for Atmospheric Research (NCAR) pioneered wind energy forecasting in 2010 with a system that saved Midwestern US utility Xcel Energy millions with three-day ahead forecasts.

NCAR is also working on a two-year plan to predict sudden changes in wind speed from severe weather events and predict output for small-scale solar energy systems, as well as a three-year project to create technology that creates three-day solar energy forecasts at 15-minute increments.

Other research initiatives have been launched to better understand how siting turbines affects wind farms and their energy output, as well as how variable renewable electricity can be better integrated into our energy system by grid operators, but results from those initiatives are years away.

More Sophisticated Analysis Than Ever Before 

While the technology race to forecast renewable energy output may be crowded, HyRef seems to have pulled ahead on two counts: the ability to forecast weather further out than any competitor, and the power of the technology in action.

HyRef system capability
HyRef system capability graphic via IBM

“Applying analytics and harnessing big data will allow utilities to tackle the intermittent nature of renewable energy and forecast power production from solar and wind in a way that has never been done before,” said Brad Gammons of IBM. “We have developed an intelligent system that combines weather and power forecasting to increase system availability and optimize power grid performance.”

Improving weather-renewables forecasting is an important imperative for the clean energy transition. The misalignment between actual renewables output and system demand stretches from up to 4 hours daily for wind to up to 1.25 hours daily for solar, according to Navigant Research, and matching renewable supply to demand could be worth up to $733 million globally.

10% Increase In Renewables Output

No other system, even those in development, have promised or delivered more than a 36-hour forecast. But HyRef can predict local weather forecasts for individual wind turbines within a wind farm and solar systems up to one-month in advance – nearly ten times the length of NCAR’s forecasts.

This long-term outlook gives grid operators unprecedented ability to plan ahead and integrate the maximum amount of clean electricity onto the grid without worrying about intermittency or forcing curtailment.

HyRef renewable energy forecast
HyRef system dashboard image via IBM

China’s State Grid Jibei Electricity Power Company Limited (SG-JBEPC) has already begun using HyRef in phase one of the 670-megawatt (MW) capacity Zhangbei wind-solar energy facility. By combining on-site energy storage with HyRef forecasts, the utility will be able to increase renewable electricity integration 10% – enough to power more than 14,000 homes compared to previous output.

Applications Beyond Renewable-Grid Integration

HyRef could revolutionize how grid operators and power developers look at renewable energy intermittency. But beyond solving intermittency challenges, HyRef may eventually help renewable energy developers find the best locations to build new projects. HyRef builds upon an IBM-Vestas project that has used big data analytics to site wind turbines based on petabytes worth of data to improve generation output and reduce maintenance and operational costs.

“Utilities around the world are employing a host of strategies to integrate new renewable energy resources into their operating systems,” said Vice Admiral Dennis McGinn of the American Council on Renewable Energy. “The weather modeling and forecasting data generate from HyRef will significantly improve this process and put us one step closer to maximizing the full potential of renewable resources.”

Check out the video below for more information about how HyRef works:


Have a tip for CleanTechnica? Want to advertise? Want to suggest a guest for our CleanTech Talk podcast? Contact us here.

CleanTechnica Holiday Wish Book

Holiday Wish Book Cover

Click to download.

Our Latest EVObsession Video

I don't like paywalls. You don't like paywalls. Who likes paywalls? Here at CleanTechnica, we implemented a limited paywall for a while, but it always felt wrong — and it was always tough to decide what we should put behind there. In theory, your most exclusive and best content goes behind a paywall. But then fewer people read it!! So, we've decided to completely nix paywalls here at CleanTechnica. But...
Like other media companies, we need reader support! If you support us, please chip in a bit monthly to help our team write, edit, and publish 15 cleantech stories a day!
Thank you!

CleanTechnica uses affiliate links. See our policy here.

4 thoughts on “HyRef Technology Revolutionizes Renewable Energy Forecasting

  • Headline: “HyRef Technology Revolutionizes Renewable Energy Forecasting”.
    Text: “HyRef joins an increasingly crowded field of innovative technologies
    seeking to accurately predict the output of renewable energy resources.”
    Which is it? Revolutions are staggeringly rare. We’ve lived through one, the arrival of the Internet. The iPod and iPhone were neat gadgets, not revolutions. I know that “another incremental improvement in one piece of the jigsaw” isn’t catchy, but it has the merit of being true.

  • Very Cool…Sometimes it takes something else to make weather forecasting a reality.

  • At least some of the claims of the article are blatantly wrong. In Denmark, we (Risø National Laboratory and DTU, now commercial as ENFOR) delivered 48-hour forecasts to end users since the early 1990ies, and week-ahead forecasts since 2004, so the claim that no-one else has delivered anything but up to 36-hour forecasts is plainly wrong.

    I am also a bit sceptical about the month-ahead forecasts – of course one can put out some numbers a month ahead, but with 15-min prediction intervals as indicated in the article, the scatter will be enormous.
    Finally, the 10% increased renewable output mentioned seem to come from the storage system compared to no storage system, not from HyRef. It would be interesting to see how much the HyRef system itself contributes.

    Having said that, the system as such involving all the big data sources looks quite promising, and I’m sure the scientists involved have not condoned the sales blurb by Silvio. I would be quite interested in learning more about the system with a bit more scientific numbers on forecasting skill etc. Maybe on the EWEA / ANEMOS Workshop on Forecasting in Rotterdam in December?

    Gregor Giebel, DTU Wind Energy (the former Risø National Laboratory)

  • ” HyRef can predict local weather forecasts for individual wind turbines
    within a wind farm and solar systems up to one-month in advance ”

    I’d need to see some data in order to believe these claims.

    If they were able to do this all our weather outlets would be using their system and telling us if Saturday, three weeks from now, was absolutely going to be a good day for a picnic.

Comments are closed.