By Johanna Koolemans-Beynen
Big data — extremely large data sets commonly analyzed by computer to reveal patterns, trends and associations — is making it possible for utilities to use more and more renewables on the grid, not only in the industrial world, but also in developing countries. That could be a game changer for global efforts to reduce emissions in both the industrial and the developing world.
Until recently, the ability of utilities to absorb electricity production from variable renewables- basically solar and wind- has been limited because of their variability. Utilities need to match electricity production very precisely with demand. They are very good at predicting what demand will be at any given time of day or year. And supply has typically been very reliable and predictable, with coal or nuclear providing the base load, and gas, oil or hydropower making up the difference by fine-tuning production to match demand.
However, incorporating renewables adds two more layers of complexity to this mix. First, the supply of energy is less easy to predict: It is hard to know when output from variable renewable energy sources might change unexpectedly due to changes in wind strength or cloud cover. Second, the increase in the use of solar panels by individual households can lead to large and unexpected increases in the demand for electricity: When cloud cover increases, these households draw more electricity from the grid.
Developing countries are already starting to see how they can use big data to integrate renewables more quickly and efficiently. For example, China, the world’s largest emitter, has committed to produce 20 percent of its primary energy consumption from renewables by 2030, but is already running into some problems while only at around 12 percent, due to mismatches between supply and demand. The World Resources Institute estimates that in China around 10 percent of solar energy, and 15 percent of wind energy, was wasted in 2015 due to this problem.
India, the world’s third-largest emitter, is beginning to face a similar situation. It is aiming to increase solar capacity from 2.9 MW in 2015 to 100 MW by 2022, with another 60 GW from wind, according to Renewable Energy World. Current renewable energy generation capacity stands at roughly 30 GW. India’s National Action Plan on Climate Change contains a target of 15 percent renewables by 2022. It is well on its way to achieving this goal, renewables having grown by 19 percent annually since 2009. However, India is already facing problems integrating the current amounts of wind and solar energy, according to Green Tech Media, which reports that this past July, the Indian state of Tamil Nadu was unable to use all the renewable energy it generated.
Careful planning will be necessary to integrate such large amounts of renewables. Big data is expected to play an important role in this expansion. Two advances in particular will expand utility capacity to integrate renewables onto the grid: increasingly complex meteorological forecasts and the Internet of Things.
Improved meteorological forecasts are expected to provide a big boost for renewables, since utilities can better estimate how much renewable energy will be available. When renewables fail to provide expected energy, utilities currently need to call on backup energy producers, often using natural gas, able to ramp up production on very short notice to meet demand. However, the Wall Street Journal reports that utilities are using big data to reduce the need for backup sources of energy, by gathering data about weather conditions directly from the solar panels and wind turbines, and combining them with data from other sources, such as weather stations, radar and satellites. This allows them to more precisely predict how much energy these sources will produce, so that they can reduce the safety margins that utilities need to cope with unexpected changes. More precise predictions will also allow utilities to better use demand management tools.
Although this is by no means a mature industry, developing countries are already starting to see how they can use big data to integrate renewables more quickly and efficiently. In India, the World Resources Institute reports that its National Institute of Wind Energy has been working with a Spanish wind modeling company, Vortex, to improve the accuracy of Tamil Nadu’s wind forecasts. And IBM developed a new technology to better predict wind energy output, in response to requests from the State Grid Corporation of China, using sensors attached to individual wind turbines together with weather information, according to website Quartz.
The Internet of Things creates opportunities on the demand-side of the equation that are at least as large. Siemens claims that improved predictions of demand via smart metering devices and other tools will help India reduce the 25 percent of electricity it loses during distribution, as well as integrate greater amounts of renewables, by better balancing supply and demand. In China, Huawei developed a wireless smart grid for the China Southern Power grid, which provides electricity to 230 million people, featuring automatic distribution and measurement, and video surveillance of the distribution network, and reduced the utility’s costs by 5 percent, according to Accenture.
Savings are likely to continue to grow as new applications for big data are developed. According to Leonida Mutuku of Intelipro, a Kenyan consultancy that builds data products and analytical tools for financial and retail organizations, "It is difficult to solve problems you can’t quantify." She says, "The use of big data enables Least Developed Countries to determine the extent of problems like food and water, energy and electricity, sanitation, and primary education before setting a roadmap to solve them."
Johanna Koolemans-Beynen is senior project coordinator at the U.S. Energy Association in Washington DC.
Return to the 2016 Winter Issue