Event Thumb

Organizer: Xu Wu, North Carolina State University

Machine Learning (ML) is a subset of Artificial Intelligence (AI) which studies computer algorithms that can improve automatically through experience (data). Deep Learning (DL) is a subset of ML that uses multi-layered neural networks to deliver state-of-the-art accuracy in tasks such as object detection, speech recognition, language translation and others. Scientific Machine Learning (SciML), more specifically, consists of computational technologies that can be trained with scientific data to augment or automate human skills. ML has been very successful in areas such as computer vision, natural language processing, etc. But its application in scientific computing is relatively new, especially in Nuclear Engineering (NE). This workshop aims at augmenting the applications of AI/ML in scientific computing in nuclear computational science, and promoting ML-based transformative solutions across various DOE missions.

Recently, ML/DL have been applied in areas such as data-driven closure model development for nuclear thermal-hydraulics, data-driven material discovery and qualification, Digital Twins for integrated energy systems, small modular reactors (SMRs) and micro-reactors, AI-based autonomous operation and control for advanced nuclear reactors, AI-based diagnosis, prognosis and predictive maintenance, etc. In this workshop, we will have five presentations that cover a wide range of topics, including:

  • Development of Neural Thermal Scattering Modules for Reactor Multi-Physics Applications, Dr. Ayman Hawari, Department of Nuclear Engineering, North Carolina State University;
  • Active Learning for Accelerating Expensive Computational Modeling Tasks, with applications on analysis on thermo-mechanical nuclear fuel failure, Dr. Som Dhulipala, Department of Computational Mechanics and Materials, Idaho National Laboratory;
  • Development of A Nearly Autonomous Management and Control System for Advanced Reactors, Dr. Linyu Lin, Department of Instrumentation, Controls & Data Science, Idaho National Laboratory;
  • Machine Learning Applications in Reactor Design and Nuclear Data, Dr. Vladimir Sobes, Department of Nuclear Engineering, The University of Tennessee, Knoxville
  • Prediction of PWR Pin Powers using Convolutional Neutral Networks, Dr. Justin Watson, Department of Materials Science and Engineering, University of Florida

Speaker Slides

Active learning for computational simulations: Application to TRISO fuel failure analysis

Development of Neural Thermal Scattering (NeTS) Modules For Data Representation and Applications

Development of A Nearly Autonomous Management and Control System for Advanced Reactors

Applications of AI/ML from Nuclear Data to Reactor Design

Prediction of PWR Pin Powers using Convolutional Neutral Networks