Monitoring power grid stability in real-time can be a cumbersome task for grid operators. An artificial intelligence (AI)-based tool has been developed by U.S. Pacific Northwest National Laboratory (PNNL) researchers to streamline decision making in this sector.

ChatGrid offers a generative AI solution to simplify and speed up decision making for U.S. power grid operators that addresses questions about generation capacity, voltage, power flow and more. After being queried, the tool generates a visualization that will show the desired information. The system runs on aSource: Cortland Johnson, PNNL publicly available large language model to search an internal database that compiles all grid infrastructure data.

The program uses synthesized data from the Exascale Grid Optimization (ExaGO) model developed by PNNL and partners. ExaGO simulates the power grid in real time, allowing grid planners to analyze the ripple effects of any disruptions. This approach enables ChatGrid to produce grid visualizations while keeping the nation’s grid data safe.

ChatGrid is available for download on GitHub in a multi-step process, but the developers are intent on formulating a one-step download route based on user feedback.