Facebook & CMU Open Catalyst Project Applies AI to Renewable Energy Storage

Original article was published by Synced on Artificial Intelligence on Medium


Facebook & CMU Open Catalyst Project Applies AI to Renewable Energy Storage

Facebook AI and the Carnegie Mellon University (CMU) Department of Chemical Engineering yesterday announced the Open Catalyst Project. The venture aims to use AI to accelerate the discovery of new electrocatalysts for more efficient and scalable storage and usage of renewable energy.

To help address climate change, many populations have been increasing reliance on renewable energy sources such as wind and solar, which produce intermittent power. The electrical energy from the intermittent power sources needs to be stored when production exceeds consumption, and returned to the grid when production falls below consumption. In California for example, solar generation peaks under the afternoon sun, while demand continues strongly into the evening.

Converting excess solar and wind energy to other fuels is a popular renewable energy storage solution, but relies on expensive electrocatalysts such as platinum for driving chemical reactions. To be widely adopted and scaled to nation-sized grids, it is necessary to find lower-cost catalysts.

Though researchers can test and evaluate new catalyst structures via quantum mechanical simulations such as density functional theory (DFT) calculations, such simulations’ high computational cost limits the number of structures that can be tested. It’s hoped the use of AI may find ways to more quickly and accurately predict atomic interactions.

The debut of the Open Catalyst Project also saw the release of the Open Catalyst Dataset for training ML models. The dataset provides 1.2 million molecular relaxations with results from over 250 million DFT calculations. have also provided baseline models and code so the broader scientific community can participate. Facebook AI and CMU believe the project could enable labs to perform days of traditional screening and calculations in just seconds.

The Open Catalyst Project and accompanying dataset can be found on GitHub.