A new solar and wind forecasting system developed by IBM researchers has proven to be 30 per cent more accurate than current technologies, a breakthrough that could have a serious impact on the integration of large-scale renewables into America’s energy grid.
The state-of-the-art system collates thousands of forecasts from national weather stations, satellite images and special sky cameras mounted at solar and wind farms. It then uses embedded machine learning to compare this data against past weather modelling to accurately predict how much renewable energy is available days – and even weeks – in advance.
IBM’s Self-learning weather Model and renewable forecasting Technology, also known as SMT, could help alleviate the issue of intermittency – the ebb and flow of power generated by weather-driven power plants. Better forecasting technology means solar and wind can become more competitive with conventional power sources because utilities do not have to hold as much energy in reserve to meet demand.
With funding from the Department of Energy’s Sunshot Initiative, the IBM team successfully tested the SMT system at ISO New England, the grid operator serving Connecticut, Maine, Massachusetts, New Hampshire, Rhode Island and Vermont.
“Solar photovoltaic resources have expanded dramatically in New England in the last five years, going from just 44 megawatts to 1,000 megawatts,” said Jonathan Black, lead engineer on ISO New England’s solar PV forecasting efforts and a collaborator in the project. “The growing aggregate output from all these resources across our region will increasingly change the daily demand curve, so the ISO will need accurate solar forecasts to help grid operators continue to balance power generation and consumer demand.”
Although solar was the second-largest source of installed energy in the U.S. in 2013 after natural gas, IBM Research Manager, Hendrick Hamann said it is only expected to provide 20-30 per cent of America’s electricity by 2050.
“However, there is good reason to believe that with better forecasts, it might be possible to push solar’s energy contribution up to 50 per cent,” Hamann said. “As we continue to refine our system in collaboration with the DOE, we hope to double the accuracy of the system in the next year. That could have a huge impact on the energy industry – and on local businesses, the economy and the natural environment.”
