Conference Tools for 2025 TMS Annual Meeting & Exhibition

Login

Register as a New User

Help

Submit An Abstract

Propose A Symposium

Presenter/Author Tools

Organizer/Editor Tools

About this Abstract
Meeting	2025 TMS Annual Meeting & Exhibition
Symposium	Artificial Intelligence Applications in Integrated Computational Materials Engineering
Presentation Title	Developing Reduced Order Models for Phase Field Modeling of Irradiation Damage Using Koopman Operator Theory
Author(s)	John Matthew Eggemeyer, Umesh Vaidya, Cheng Sun
On-Site Speaker (Planned)	John Matthew Eggemeyer
Abstract Scope	Irradiation damage in materials is a significant concern for the safety and reliability of nuclear reactors. Phase-field models offer a versatile framework for simulating irradiation damage at mesoscales, such as forming fission gas bubble superlattices, which occur under specific irradiation conditions (dose, dose rate, and temperature). However, these high-fidelity models are computationally expensive. To address this, Koopman operator theory is employed to develop reduced-order linear models, enabling instantaneous simulations of fission gas bubble behaviors. We apply the Dynamic Mode Decomposition (DMD) algorithm to derive these linear Koopman models from simulation data. These low-fidelity linear models predict the formation window of gas bubble superlattices and are validated against high-fidelity phase-field models and experimental results. This paper presents our findings on using phase-field models of irradiation damage, along with the DMD method, to reduce computational costs while maintaining the accuracy of high-fidelity simulations in predicting irradiation-induced microstructures.
Proceedings Inclusion?	Planned:
Keywords	Computational Materials Science & Engineering, Nuclear Materials, Machine Learning

OTHER PAPERS PLANNED FOR THIS SYMPOSIUM

A Bayesian Approach for Constitutive Model Selection and Calibration Using Diverse Material Responses

A Machine Learning Informed Phase Field Damage Model to Simulate Void Nucleation and Growth in Metal Microstructures

A Multiscale Simulation Framework for Incremental Deformation Processing Using a Recurrent Neural Network Surrogate Model for Crystal Plasticity

A Study on the Smoke Recognition of Steelmaking Plants Based on EL-MobileNet

Accelerating Crystal Plasticity Simulations with Graph Neural Networks

AI in ICME: Methodologies for AI Alignment and Explainability in Self-Driving Labs

ANNA: An Open-Source Platform for Developing Artificial Neural Networks Assistant Potential Enabling High Accurate and Efficient Molecular Dynamics Simulation

Automation of the ICME Workflow Incorporating Material Digital Twins at Different Length Scales Within a Robust Information Management System

Combined THz-TDS and Raman Spectroscopy for In-Situ Material Identification via a Machine Learning Algorithm

Conditional Diffusion Models for Interlocking Metasurface Design

Data-Driven Modeling of Dislocations for Multi-Scale Simulations

Data and Decision Science-Driven Assessment and Selection of Mg Alloys for Fracturing Applications

Data Assimilation of Multi-Phase-Field Model based on Physically Informed Neural Network

Data Modelling of Through-Life Structural Integrity Assessment of Dissimilar Metal Welds for Nuclear Application

Design of High-Strength Steel Using Machine Learning Techniques

Developing a Foundational Inter-Atomic Potential for Transitional Metal Alloys Using Active Learning

Developing Machine Learning Interatomic Potential for Fe-Cr-Ni Alloys

Developing Reduced Order Models for Phase Field Modeling of Irradiation Damage Using Koopman Operator Theory

Digital Twins for Accelerated Materials Innovation

Effect of the Microstructure on Intergranular Fracture in FCC and HCP Polycrystals: A Machine Learning Approach

Enhancing Extrusion Efficiency: Development of a Digital Twin for Glass Reinforced Polymer Processes Using Machine Learning and Real-Time Data Integration

Enhancing Medical Waste Recycling Through Computer Vision and Near-Infrared Spectroscopy

Establishing a Novel Systematic Alloy Design Strategy Based on Large Language Model Framework

Generative Adversarial Network (GAN)-Based Microstructure Mapping from Surface Profile For Laser Powder Bed Fusion (LPBF)

Harnessing Graph Neural Networks for Classification of Unique Glassy Structures in CuZr Metallic Glasses

High-Throughput and Robust Materials Design Hypothesis Generation via a RAG-Enhanced Large Language Model

Machine Learning-Driven Multiscale Analysis of Mechanical Properties in Metal-Matrix Nanocomposites

Machine Learning and High-Throughput Computations Guided Development of High Temperature Oxidation-Resisting Ni-Co-Cr-Al-Fe High-Entropy Alloys

Machine Learning Facilitated Integration of Characterization Data and Simulations to Generate Residual Stress Distributions

Machine Learning Potentials and Other Tools in LAMMPS for Materials Engineering

Magnetic RANN Interatomic Potential for Iron

Physical Metallurgy and Machine Learning Guide the Prediction of Continuous Cooling Phase Transformation in Steels

Prediction of Fatigue Indicator Parameter by Graph Neural Network

Prediction of Material Parameters Using Machine Learning Supported by Large-Scale Phase-Field Simulations of Dendrite Growth

Pushing the Limits of Fine Feature Detection in Deep-Learning Assisted 3D X-Ray Microscopy: Characterization of Hierarchical Microstructures in TiC Reinforced Nickel Matrix Composites

Rapid Microstructural Determination from Nano-indentation of High Entropy Alloys Using Machine Learning and Genetic Algorithms

Starrydata Explorers: Visualization Platforms to Overview the Past Reported Experimental Samples

Sustainable Aluminum Alloy Design via Computer Vision

Towards Automatic Alloy Design via Large Language Model Powered Multi-Agent Collaborations

Tuning Fracture Characteristics for Chiral Aperiodic Monotile Based Composites by Employing Multi-Objective Bayesian Optimization

Questions about ProgramMaster? Contact programming@programmaster.org