AI-Driven dynamic radiological source-term model for radiation protection in D&D

AI-Driven dynamic radiological source-term model for radiation protection in D&D: Enhancing Safety and Efficiency through Digital Twin Integration

In nuclear decommissioning, digital twins are increasingly used. They can offer an improved efficiency in the decommissioning procedures, and will help in informed decision making towards safety of the workers during operations such as cutting pipes and handling radioactive materials. Accurately knowing the radioactive source term is crucial for the usefulness of the digital twin. Therefore this source term must be dymically updated throughout the decommisioning activities. This PhD topic aims to revolutionize this process by employing AI and machine learning to develop a predictive, adaptive model.

The core of this PhD research revolves around the integration of digital twins with the radiation protection and ALARA procedures. The primary research question investigates whether AI/ML models can effectively track and predict changes in the radioactive source term. This innovative approach seeks to improve traditional, labor-intensive measurement methods with a system that can dynamically anticipate changes in the radioactive environment. Of course measurements will still be needed, but the amount and timing of these measurements should be optimized.

As a PhD candidate, your objective will be to create AI/ML models that intricately process gamma detection data and historical operational procedures. Your research will empower Digital Twins, ensuring they evolve continuously, becoming invaluable assets in informed decision-making during nuclear decommissioning.

Be a part of this exciting and innovative journey, where your research will pave the way for safer, more efficient nuclear decommissioning practices.

Estimated duration