arXiv:2607.04450v1 Announce Type: cross Abstract: Integral codes like the Accident Source Term Evaluation Code (ASTEC) are powerful tools to study the physics of Severe Accidents (SAs) in nuclear reactors. Real time SA simulators can also be helpful in training operators of nuclear plants to react correctly to malfunctions. However, SA simulators can take up to several days per simulation, making their use infeasible for real time applications. In this work we show how to speed up a SA simulator with a fast, Deep Learning based (DL), surrogate model (SM). The SM is built as a combination of a
Advances in Deep Learning coupled with the increasing need for operational efficiency and safety in critical infrastructure sectors make this development timely.
This development allows for real-time simulation of complex nuclear accident scenarios, which is critical for operator training and improving safety protocols in nuclear power plants.
Real-time, high-fidelity simulation of severe nuclear accidents, previously infeasible due to computational demands, becomes a reality through AI-driven surrogate models.
- · Nuclear power industry
- · Nuclear plant operators
- · AI/Deep Learning companies
- · Safety regulation bodies
- · Traditional simulation software vendors (if they don't adapt)
- · Facilities relying solely on slow, legacy simulation methods
Operators can train more effectively and respond to emergencies with greater precision, reducing the risk of severe incidents.
The cost and complexity of safety training for nuclear facilities may decrease, potentially encouraging broader adoption of nuclear energy.
This methodology could be adapted for real-time risk assessment and operational control in other complex, high-stakes infrastructure, like chemical plants or space missions.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.AI