Description

Fast breeder reactors are next-generation nuclear reactors that not only generate power but also breed fuel, making them highly efficient and sustainable. It uses liquid metallic sodium as a coolant due to its superior thermal conductivity and neutronic properties. However, sodium’s high reactivity with air and water poses significant safety challenges
that impact reactor systems and components.
This talk explores how advanced AI techniques are transforming these risks into robust safety solutions, enhancing reactor resilience and operational reliability. We will present key AI-driven applications including early detection of hydrogen explosions during sodium fires, rapid identification of sodium leaks, predictive maintenance strategies for critical reactor subsystems, and autonomous fire suppression systems. All aimed at elevating nuclear safety standards.

Outline of Talk

1. Fast Breeder Reactors

• Overview of three stage nuclear program with a focus on the role of fast breeder reactors and the challenges associated with it.

2. AI Use Cases for Safety Enhancements

2.1 Detection of Hydrogen Explosion in Sodium Fire

• For the safe disposal of used radioactive sodium, a spray injection technique is employed.
• Hydrogen is generated during this process, and uncontrolled spraying can trigger explosions.
• AI models detect early warning signs of hydrogen accumulation or ignition risk, allowing timely control to prevent catastrophic events.

2.2 Sodium Leak Detection at Early Stage

• Leakage of sodium from the reactor containment area leads to sodium aerosol (smoke) generation.
• Sensor based real-time monitoring tracks the aerosol formation.
• Machine learning algorithms analyze aerosol patterns to detect leaks early and enable swift corrective action.

2.3 Predictive Maintenance of Reactor Subsystems

• Reactors have stringent operational constraints for shutdown and restart.
• Although subsystems are redundant and regularly maintained, predictive analystics models forecast component life and detect anomalies before failure.

2.4 Automated monitoring of operator safety

• A computer vision system continuously verifies whether operators are wearing required safety gear like helmets, TLDs (Thermoluminescent Dosimeters), and jackets.
• AI-powered monitoring ensures compliance and triggers real-time alerts for violations, thereby reducing human error in safety-critical environments

Future Work

• Integration of Generative AI for enhancing safety applications

Speaker Info:

Muthukumar Ganesan is currently working as a Scientist at the Atomic Research Centre, Government of India. He brings over 11 years of professional experience in AI application development across the automotive and nuclear industries. His work focuses on enhancing safety and automation in sodium-cooled fast breeder reactors using AI-driven solutions, including hydrogen explosion detection, sodium leak identification, predictive maintenance, and autonomous fire mitigation systems. He has published 3 research papers in the field of AI and is currently working on 2 more. An experienced technical speaker, he has delivered talks on AI applications in industrial systems and actively contributes to the GitHub and MathWorks communities, where he shares innovative tools and solutions for real-world engineering challenges.

Linkedin | Git | Mail