Machine learning techniques for anomaly detection and for the alignment of perturbation simulations with power plant measurements

Main Objectives

Detect anomalies in nuclear reactors using non‐intrusive methodologies
Anomalies
• Excessive vibrations of core internals
• Flow blockage
• Coolant inlet perturbations
• Combination of the above
• …
Non‐intrusiveness
• Measure the inherent fluctuations in neutron flux recorded by in‐core and ex‐core
detectors
• No external perturbation of the system is required

More information can be found in the presentation file