System automation

Argonne National Laboratory Creating AI-Powered Systems for Low-Cost Nuclear Energy

The Department of Energy’s Argonne National Laboratory is developing artificial intelligence-enabled systems that could help produce nuclear energy at a lower cost compared to existing nuclear power plants.

Work is being performed under a $1 million, three-year project funded by the DOE Office of Nuclear Energy’s Nuclear Energy Enabling Technologies program.

The goal is to build smart, computerized systems that can enable early problem detection and provide human operators with recommendations to resolve the issue, Argonne National Laboratory said.

Power plants require large crews and extensive maintenance, resulting in large operation and maintenance costs. According to Roberto Ponciroli, a principal nuclear engineer at Argonne, autonomous operations can help improve profitability and accelerate the deployment of advanced reactor concepts. The technology is expected to deliver over $500 million in savings annually.

AI can also help verify data from hundreds of sensors to ensure proper monitoring of power plant systems. “In a world where decisions are made according to data, it’s important to know that you can trust your data,” Ponciroli said. The official noted that sensors can degrade over time and inspection of the equipment is currently performed manually. With the planned solution, algorithms will be used to validate sensor data by learning how a normal sensor works.

Experts at Argonne have partnered with the industry to create a digital twin of an advanced nuclear reactor to test the AI technology’s capability. The laboratory has already completed tests focused on problem detection. Trials will soon evaluate how the smart system will make decisions based on gathered data.