Argonne National Laboratory develops AI sensor system for nuclear reactors

Researchers at the US Department of Energy’s Argonne National Laboratory have developed a novel monitoring system designed to detect hidden blockages within nuclear reactor heat exchangers. The technology combines distributed fiber-optic temperature sensors with artificial intelligence to monitor thousands of individual channels in real time.

The system is specifically intended for use in molten-salt-cooled reactors (MSCRs). In these reactors, hot salt transfers thermal energy from the core to another fluid to generate electricity. If the temperature drops near the freezing point, the salt can solidify and block the small channels within the heat exchanger, potentially leading to costly failures and reactor shutdowns.

While traditional monitoring typically only tracks temperature and flow at the inlet and outlet, this new method embeds sensors into the support structures of the heat exchanger. AI algorithms then analyze the massive amounts of data collected to identify unusual temperature patterns, allowing operators to detect, locate, and assess the severity of channel blockages before they cause significant issues.