US lab adapts ChatGPT-like transformer tech to speed up nuclear reactor simulations

Argonne National Laboratory is adding transformer-based AI to its SAM simulation tool to speed up modeling of turbulence and heat transfer in advanced nuclear reactors.

The goal is to combine AI speed with the accuracy of physics-based methods for better reactor simulation.

Researchers are also testing the approach for whole-plant simulations and digital twins.

If successful, it could improve design, safety analysis, and monitoring while lowering computing costs.