Google Cloud today announced plans to develop an artificial intelligence (AI) and data management platform designed to optimize renewable wind energy and render it more cost effective than carbon-intensive fossil fuels.

Developed by Google Cloud's AI services and industry solutions (AIIS) team, the platform will be used by energy provider ENGIE to predict wind patterns and optimize its portfolio of wind energy projects in Germany.

The goal of the pilot is to predict how much wind power should be sold on which power market and at what price, according to the cloud giant. On a broader scale, the two companies say this project will enable transactions for wind asset developers and add value for wind energy providers, thereby accelerating the clean energy transition and global grid decarbonization.

The complexity of short-term power markets and the unpredictability of wind availability are among the biggest hurdles facing broader wind energy generation and adoption. Large amounts of data from different sources need to be gathered, stored, and analyzed to optimize energy production.

Google claims its platform will be able to use a performant and scalable data system and machine learning algorithms "to extract value from the data that supports subsequent decisions."

“At Google Cloud, we believe that more accurate data and predictions of wind power production will be valuable to electricity grids, creating benefits for consumers and making wind more competitive with fossil fuels," Larry Cochrane, director of Google Cloud's global energy solutions, said in a statement.

He added that working with ENGIE will accelerate Europe's transition to renewables and lay the groundwork for other wind energy providers to utilize AI forecasting.

This partnership marks the platform's debut, though Google plans to extend the platform to additional energy providers in the future, citing hundreds of gigawatts of wind energy farms globally that would be able to optimize energy generation with advanced AI forecasting.

Although this is the first time this solution will be deployed with an energy provider partner, Google has previously used similar predictive tech in its own wind portfolios.

Using a neural network trained on weather forecasts and historical wind turbine data, Google says it's able to predict wind power output 36 hours before generation. And based on those predictions, its model recommends optimal hourly delivery commitments to the power grid up to a day ahead of time.

This means energy sources can be scheduled to deliver a certain amount of electricity at a certain time, which is more valuable to the electricity grid than traditional wind energy generation processes.

According to the provider, machine learning has increased the value of its wind energy by roughly 20% compared to the baseline scenario of no time-based commitments to the grid.

The ENGIE partnership also directly supports Google's commitment to decarbonize its global operations.

The cloud giant plans to power its data centers and offices with continuous carbon-free energy (CFE) by 2030, meaning it will match each hour of its electricity consumption with CFE on every grid where the provider operates. Google posits CFE as crucial both to itself and to the global economy, and it encourages companies to go beyond simply buying clean energy in the form of power purchase agreements, for example.

“To achieve 24/7 CFE, both for our own operations and for economies more broadly, we must do more than just purchase clean energy — we must work to transform electricity systems,” Caroline Golin and Nick Pearson, who lead Google’s energy market development and policy, explained in a blog post.