AI Nuclear Reactor Design Research

Machine Learning for Neutron Transport Simulation

Interns will develop and train neural network models to accelerate neutron transport calculations in reactor core simulations. They will work with existing Monte Carlo and deterministic solvers, creating surrogate models that reduce computational time while maintaining accuracy for reactor design optimization.

AI-Driven Reactor Core Configuration Optimization

Interns will apply reinforcement learning and genetic algorithms to optimize fuel assembly arrangements, control rod positioning, and core geometry for enhanced safety and efficiency. This work involves parameterizing reactor designs and developing objective functions for multi-criteria optimization.

Anomaly Detection in Reactor Thermal-Hydraulics Data

Interns will build machine learning models to identify abnormal patterns in reactor cooling system data, temperature distributions, and pressure profiles from simulation and operational datasets. They will focus on developing real-time monitoring algorithms for early detection of potential system degradation.

Deep Learning for Radiation Shielding Design

Interns will use convolutional neural networks and physics-informed neural networks to predict radiation dose distributions and optimize shielding material configurations around reactor vessels. The research will involve training models on MCNP simulation data and validating against experimental benchmarks.

Generative AI for Novel Reactor Concept Development

Interns will explore generative models and topology optimization techniques to design innovative reactor core architectures and component geometries that meet safety and performance constraints. They will integrate domain knowledge from nuclear engineering with AI-generated designs to propose feasible advanced reactor concepts.