arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2602.19915cs.LG
收藏

全卷积时空学习用于微结构演化的预测

Fully Convolutional Spatiotemporal Learning for Microstructure Evolution Prediction

Department of Mathematics and Systems Engineering(数学与系统工程系);Florida Institute of Technology(佛罗里达理工学院);Department of Mathematics(数学系);Embry-Riddle Aeronautical University(埃姆布里-瑞德尔航空大学);Department of Mechanical and Civil Engineering(机械与土木工程系)

Michael Trimboli, Mohammed Alsubaie, Sirani M. Perera, Ke-Gang Wang, Xianqi Li

AI总结:

本文提出了一种全卷积时空模型,用于高效预测微结构演化,相比传统方法具有更高的准确性和更低的计算成本。

摘要:

理解并预测微结构演化是材料科学的基础,因为它决定了材料的最终性能和特性。传统模拟方法,如相场模型,能够提供高保真的结果,但因为需要求解复杂的偏微分方程,所以在细粒度时空分辨率下计算成本很高。为了解决这一挑战,我们提出了一种基于深度学习的框架,以加快微结构演化预测,同时保持高精度。我们的方法利用了一个完全卷积的时空模型,该模型通过自监督的方式使用由微结构过程模拟生成的序列图像进行训练,包括晶粒生长和分形分解。训练好的神经网络能够有效学习底层的物理动态,并且能够准确捕捉演化的微结构的短期局部行为和长期统计特性,同时还能推广到未见的时空领域和配置及材料参数的变化。与循环神经网络架构相比,我们的模型在训练和推理中都实现了显著降低的计算成本,预测性能达到最先进的水平。本工作为材料科学中的时空学习建立了稳健的基准,并提供了一种可扩展的数据驱动替代方案,用于快速可靠地进行微结构模拟。

英文摘要:

Understanding and predicting microstructure evolution is fundamental to materials science, as it governs the resulting properties and performance of materials. Traditional simulation methods, such as phase-field models, offer high-fidelity results but are computationally expensive due to the need to solve complex partial differential equations at fine spatiotemporal resolutions. To address this challenge, we propose a deep learning-based framework that accelerates microstructure evolution predictions while maintaining high accuracy. Our approach utilizes a fully convolutional spatiotemporal model trained in a self-supervised manner using sequential images generated from simulations of microstructural processes, including grain growth and spinodal decomposition. The trained neural network effectively learns the underlying physical dynamics and can accurately capture both short-term local behaviors and long-term statistical properties of evolving microstructures, while also demonstrating generalization to unseen spatiotemporal domains and variations in configuration and material parameters. Compared to recurrent neural architectures, our model achieves state-of-the-art predictive performance with significantly reduced computational cost in both training and inference. This work establishes a robust baseline for spatiotemporal learning in materials science and offers a scalable, data-driven alternative for fast and reliable microstructure simulations.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑