Google reportedly used an AI software developed by its subsidiary, DeepMind, to predict the output of its wind farms in the central United States.
According to DeepMind’s Project Manager Sims Witherspoon and Google’s Carbon Free Energy Program Lead Will Fadhonc, the predictions have helped the company optimize the hourly delivery of wind energy to the power grip one day in advance.
While wind farms have become a significant source of carbon-free, sustainable energy in the past decade, the nature of wind made them an unpredictable source of power which lessens their usefulness.
As a solution, Google, which is among the many multinational companies turning to renewable energy to power their facilities, collaborated with DeepMind to apply machine learning algorithms to 700 megawatts of wind power capacity in its wind farms.
In its statement, Google said:
“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.”
Predicting Power Output Through an AI Software
The predictions made by the AI software used by Google helped it to schedule the power delivery of the wind farms to the power grid. The company claimed that while they are still refining their algorithm, it has already produced positive outcomes.
Google said that the value of their wind energy has already increased by 20 percent since they applied the machine algorithms a year ago.
“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.”
Aside from adding value to the energy generated by the wind farms, the algorithms also enable the wind farm operators to make faster, smarter, and more data-driven decisions to meet electricity demand.
Google now hopes its machine learning algorithms could help strengthen the business case of wind power, encouraging other companies to use carbon-free energy on electric grids across the globe.
Comments (0)
Least Recent