Proving the Business Case for the Internet of Things

Machine learning helps IBM boost solar and wind energy resources

Steve Rogerson
July 28, 2015
 
Solar and wind forecasts using machine learning and other cognitive computing technologies are proving to be as much as 30 per cent more accurate than ones created using conventional approaches, according to IBM Research. Part of a research programme funded the by the US Department of Energy’s SunShot initiative, the breakthrough results suggest ways to optimise solar resources as they are increasingly integrated into the nation’s energy systems.
 
IBM also announced that for a limited time it will provide foundational solar forecasts at 5km spatial resolution to help government agencies and other organisations in the lower 48 states best evaluate their impact on supply and demand as well as operations.
 
Entering the third year, IBM researchers worked with academic, government and industry collaborators to develop a self-learning weather model and renewable forecasting technology, known as SMT. The SMT system uses machine learning, big data and analytics continuously to analyse, learn from and improve solar forecasts derived from a large number of weather models. In contrast, most current forecasting techniques rely on individual weather models that offer a more narrow view of the variables that affect the availability of renewable energy.
 
“By improving the accuracy of forecasting, utilities can operate more efficiently and profitably,” said Bri-Mathias Hodge, who oversees the Transmission & Grid Integration Group at the National Renewable Energy Laboratory (NREL), a collaborator in the project. “That can increase the use of renewable energy sources as a more accepted energy generation option.”
 
IBM’s approach provides a general platform for renewable energy forecasting, including wind and hydro. It advances the state-of-the-art by using deep machine learning techniques to blend domain data, information from sensor networks and local weather stations, cloud motion physics derived from sky cameras and satellite observations, and multiple weather prediction models. The SMT system is said to be the first time such a broad range of forecasting methods have been integrated onto a single, scalable platform.
 
“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 Siyuan Lu, physical analytics researcher at IBM.
 
In 2013, solar was the second-largest source of new electricity generating capacity in the USA, exceeded only by natural gas. A US SunShot vision study suggests that solar power could provide as much as 14 per cent of US electricity demand by 2030 and 27 per cent by 2050.
 
Currently, there are two main customers for renewable energy forecasting technologies – utility companies and independent system operators (ISOs). However, the inherent difficulty in producing accurate solar and wind forecasts has required electric utilities to hold higher amounts of energy reserves as compared with conventional energy sources. With solar power installations rapidly growing, future solar penetration levels will soon require increased attention to the value of more accurate solar forecasting.
 
“Solar photovoltaic resources have expanded dramatically in New England in the last five years, going from just 44 to 1000MW,” said Jonathan Black, lead engineer on ISO New England’s solar PV forecasting efforts and a collaborator in the project. “Currently, most of the solar installations in New England are behind the meter on the distribution system, so their output isn’t visible in real time to the ISO’s system operators, but it reduces the amount of electricity demand they observe. 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.”
 
The US Department of Energy SunShot initiative is a collaborative national effort that aims to drive innovation to make solar energy fully cost-competitive with traditional energy sources before the end of the decade.
 
Now in its 70th year, IBM Research has more than 3000 researchers in 12 laboratories across six continents.