The unpredictability of renewable energy sources like wind and solar creates reliability challenges for utilities seeking to balance power supply and demand across centralized grid networks. Policies compelling energy providers to incorporate more renewable energy into their portfolios make the challenge more urgent, but also create business opportunities for companies finding solutions.
Those who crack the utility-scale energy storage code will not only permanently transform the energy delivery system model, but make billions of dollars doing so. In the meantime, however, other solutions will need to fill the gaps between intermittent generation and base load power. IBM’s “Hybrid Renewable Energy Forecasting” (HyRef) technology could be such a solution.
The tool uses a combination of advanced weather forecasting systems – like advanced cloud imaging technology and sky-facing cameras – and big data to accurately predict weather conditions in very specific locations, like wind farms or utility-scale solar operations. The application can focus down to single wind turbine granularity.
“HyRef uses big data and analytics that simulate and forecast when will these units [panels or turbines] perform and how,” Michael Valocchi, Vice-President/Partner Global Energy and Utilities Leader, IBM Global Business Services recently explained to Breaking Energy.
“The technology achieves predictability within 10% from a forecasting perspective,” he said. Meaning accuracy within 10% of what you predict the output of a given wind farm or solar installation will be in the future.
“You can go day or week ahead with traditional power plants, now we can do day-to-day output predictions of wind and solar so you know how much backup might be needed,” Valocchi said.
A subsidiary company of the State Grid Corporation of China (SGCC) is currently using HyRef technology to integrate more renewable energy into the power grid, which is the first phase of the Zhangbei 670MW demonstration project, the world’s largest renewable energy initiative combining wind and solar power, energy storage and transmission, according to IBM.
HyRef was developed by researchers at IBM’s labs in China and the TJ Watson Research Lab in Yorktown, NY.
While only deployed in China right now, “the solution is applicable across the globe and in different markets,” said Valocchi.
“There’s interest in different parts of the value chain. There is interest from utilities who care about reliability. Equipment providers might also be interested, but utilities are really the focus,” he explained.
The technology builds on advanced weather analytics like the company’s Deep Thunder initiative – a HyRef input – and smarter analytics work developed in Denmark with Vestas. IBM is finding that as it develops technology for a given application it can increasingly be applied to solutions in other areas. “You see an advancement of the technology for more and more and different uses,” Valocchi said.
“We see this as an important component as we move toward greater percentages of renewable energy in the grid. These analytics and advanced modeling help drive the operational benefits [of renewables], which is good because policy is driving increased renewable energy usage.”
