DOE says new software could be rolled out across all nation’s BWR units.

US-based technology company Blue Wave AI Labs has successfully deployed machine learning (ML) tools at two nuclear power plants operated by Constellation, potentially saving the company millions of dollars per reactor each year, the US Department of Energy said.

The project was part of a $6m (€5.5m) effort supported by the DOE to help lower the operating costs of nuclear power plants using the latest artificial intelligence (AI) and ML technologies.

Two US national laboratories — Argonne and Brookhaven — contributed to the project.

Blue Wave projects that the new software could save up to $80m per year once the tools are expanded to the nation’s fleet of 32 boiling water reactors (BWR).

Blue Wave tested its technology at Constellation’s Peach Bottom and Limerick nuclear power stations starting in 2022.  Peach Bottom and Limerick, both in Pennsylvania, have two operational BWR each.

Blue Wave’s AI tools used vast amounts of historical plant data to analyse and improve sensor measurements within the reactor core.

Reactor operators depend on sensors to measure power generation, fuel consumption, and the overall state of the reactor with respect to operating limits.

Over time, these sensors can become out of calibration and lose accuracy. If enough sensors stop working correctly, the reactor will reduce power or shut down as a precautionary measure, costing an operator millions of dollars per day in lost generation revenue.

The DOE said that in 2023, Blue Wave identified sensors at the Limerick-2 BWR that were suspected to be out of calibration. These sensors were taken offline, allowing the plant to continue operating.

During the next sensor calibration cycle, plant operators were able to verify that sensors that were taken offline were giving incorrect readings due to miscalibration, as was predicted by Blue Wave’s tool.

The DOE said the AI algorithms also improved engineers’ ability to predict how much fuel must be purchased and how to configure the fuel to generate the greatest amount of power while preserving margin to operating limits – another time-consuming and expensive process.