Google uses DeepMind to improve wind power by 20%

DeepMind's machine-learning algorithms enable scheduling of set deliveries of energy output from wind turbines


Google has announced that it has begun using AI software developed by its London-based AI subsidiary DeepMind to ensure the energy produced by its wind farms is up to 20% more viable. By applying DeepMind's machine-learning algorithms, the tech giant said it was able to schedule set deliveries of energy output.

Renewable energy helps to combat climate change, but many renewable energy technologies have not reached their full potential yet, Google noted. Wind power, in particular, is an unpredictable power source compared to a source that can deliver power at a set time. However, by using DeepMind AI, the company has claimed that the value of the energy its farms are outputting is now 20% more than baseline.

"Using a neural network trained on widely available weather forecasts and historical turbine data, we configured the DeepMind system to predict wind power output 36 hours ahead of actual generation," explained DeepMind program manager Sis Witherspoon and Will Fadrhonc, carbon free energy program lead at Google, in a blog post.

"Based on these predictions, our model recommends how to make optimal hourly delivery commitments to the power grid a full day in advance. This is important, because energy sources that can be scheduled (i.e., can deliver a set amount of electricity at a set time) are often more valuable to the grid.

"We can't eliminate the variability of the wind, but our early results suggest that we can use machine learning to make wind power sufficiently more predictable and valuable," they added.

DeepMind has been involved in Google's renewable goals before, for example in 2016 its AI was used to cut the power costs of its data centers by 15%.

Last year, Google revealed that it now offsets its energy usage with 100% renewable sources.

Making sense of the confusing world of iot platforms small

Read next:

Making sense of the confusing world of IoT platforms