DeepMind trains the AI system on past weather forecast and histories. (Photo via pixabay)
Wind power is a popular option when it comes to clean, renewable energy. But it’s also finicky since wind is often unpredictable making it difficult to use for power grids. Google believes they may have solved that problem by using artificial intelligence. The company announced they have trained an AI system how to predict the energy output of Google’s wind farms by using DeepMind.
Google and DeepMind trained the system using widely available weather forecast and historical turbine data to help it produce wind power predictions 36 hours ahead of actual power generation. Using these predictions, the system will recommend how to make optimal hourly delivery commitments to the power grid a full day in advance. Though they’re still refining the system, Google claims DeepMind’s AI has increased the “value” of its wind energy by 20 percent.
“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,” write Sims Witherspoon, a product manager at DeepMind, and Will Fadrhonc, Google’s Carbon Free Energy program lead, in a co-authored blog post. “This approach also helps bring greater data rigor to wind farm operations, as machine learning can help wind farm operators make smarter, faster and more data-driven assessments of how their power output can meet electricity demand.”
This isn’t the only time Google has used DeepMind to conserve energy. In 2016, the company successfully cut down on the power costs of their data centers by 15 percent using the AI system. DeepMind is also in talks with the UK’s National Grid about using its AI technology to better predict energy demands.
AI is being used more to aid and improve renewable energy resources. GEV Wind Power, one of the largest wind operations and maintenance companies in Europe, recently partnered with computer vision solutions company Clobotics to use AI for wind turbine blade inspections. An autonomous drone will fly around the turbine and build a 3D picture of all the blades, capturing high-resolution images as it flies. An inspection that used to take hours is done in a matter of minutes. The pictures are automatically analyzed, annotated, and reported on the cloud-based customer portal the next day. Customers can then use the portal the track the turbines, assess any damage, and see trend reports.
These are examples of how machine learning can be used to solve critical problems, like renewable energy. The better it gets, the more we’ll see companies using it for their own energy needs. At least the technology is being put to good use instead of just beating up professional StarCraft II players and predictive text.
Have a story tip? Message me at: cabe(at)element14(dot)com