Google has announced on its company blog that it has successfully managed to boost the value of its generated wind power by 20%, thanks to installing artificial intelligence (AI) algorithms in some of its facilities.
Wind power has been consistently evolving in recent years thanks to improvements in production costs and operational efficiency, as well as advanced lubricants like Mobil SHC Gear 320 WT from Mobil UK stockists, but the successful application of machine-learning algorithms should further bolster the business case for wind power.
Like many other tech companies, Google has been engaged in moving its power-hungry data centres to renewable energy, and since April 2018, it has been powering all of its operations with renewable energy.
Last year, Google decided to implement a machine-learning algorithm developed by Deep Mind (a subsidiary of Alphabet Inc., the parent company of Google) to 700 MW of its wind power capacity. The algorithm predicts the output of windfarms based on weather forecasts and anticipates energy price fluctuations 36 hours in advance. It then makes recommendations about making optimised delivery commitments to the grid a full day beforehand. Energy that can be predictably delivered at a set time is more valuable to the grid, and this meant that Google’s produced energy was 20% more valuable than it would have been if no commitments were made.
Sims Witherspoon, DeepMind’s programme manager, said about the news:
“Our hope is that this kind of machine learning approach can strengthen the business case for wind power and drive further adoption of carbon-free energy on electric grids worldwide.”