Predicting Grain Growth in Polycrystalline Materials Using Deep Learning Time Series Models

2025-11-07

AI Summary

Predicting grain growth evolution in polycrystalline materials entails high computational cost and suffers from poor long-term modeling stability. To address this, we propose a time-series modeling framework based on low-dimensional physics-informed statistical descriptors and Long Short-Term Memory (LSTM) networks. Instead of computationally expensive full-field simulations, our approach constructs a lightweight, interpretable, and physically consistent recursive model using mean-field statistical features to enable efficient dynamic prediction of grain size distribution. Compared with alternative architectures—including RNNs, Temporal Convolutional Networks (TCNs), and Transformers—our LSTM-based method achieves superior accuracy (>90%), drastically reduced inference time per sequence (from 20 minutes to several seconds), and enhanced long-horizon prediction stability. This work establishes a scalable, physics-aware paradigm for digital twin construction of polycrystalline materials and real-time optimization of thermomechanical processing.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: Learning & Optimization for NLP
• Computer Vision: Low Level & Physics-based Vision

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Web Mining and Content Analysis: Models for Web evolution
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Grain Growth strongly influences the mechanical behavior of materials, making its prediction a key objective in microstructural engineering. In this study, several deep learning approaches were evaluated, including recurrent neural networks (RNN), long short-term memory (LSTM), temporal convolutional networks (TCN), and transformers, to forecast grain size distributions during grain growth. Unlike full-field simulations, which are computationally demanding, the present work relies on mean-field statistical descriptors extracted from high-fidelity simulations. A dataset of 120 grain growth sequences was processed into normalized grain size distributions as a function of time. The models were trained to predict future distributions from a short temporal history using a recursive forecasting strategy. Among the tested models, the LSTM network achieved the highest accuracy (above 90%) and the most stable performance, maintaining physically consistent predictions over extended horizons while reducing computation time from about 20 minutes per sequence to only a few seconds, whereas the other architectures tended to diverge when forecasting further in time. These results highlight the potential of low-dimensional descriptors and LSTM-based forecasting for efficient and accurate microstructure prediction, with direct implications for digital twin development and process optimization.

Problem

• Predicting grain size distributions in polycrystalline materials using deep learning
• Evaluating RNN, LSTM, TCN and transformer models for microstructure forecasting
• Developing efficient alternatives to computationally demanding full-field simulations

Innovation

• Using deep learning time series models for grain growth
• Employing mean-field statistical descriptors from simulations
• LSTM network achieves high accuracy and fast prediction