数据与计算发展前沿 ›› 2026, Vol. 8 ›› Issue (5): 61-71.

CSTR: 32002.14.jfdc.CN10-1649/TP.2026.05.005

doi: 10.11871/jfdc.issn.2096-742X.2026.05.005

• 专刊：“东方”超级计算系统，助力青年科学家科技创新 • 上一篇    下一篇

计算机辅助玻璃材料性能预测的研究现状及展望

杨宇帆1,2(),李东祥1,*(),周鹏2,*()   

  1. 1 青岛科技大学， 化学与分子工程学院， 山东 青岛 266042
    2 中国科学院过程工程研究所， 生物药制备与递送全国重点实验室， 北京 100190
  • 收稿日期:2026-03-19 出版日期:2026-10-20 发布日期:2026-10-20
  • 通讯作者: *李东祥（E-mail: lidx@qust.edu.cn）；周鹏（E-mail: zhpayne@ipe.ac.cn）
  • 作者简介:杨宇帆，青岛科技大学和中国科学院过程工程研究所联合培养硕士研究生。目前主要从事机器学习与分子模拟在材料科学领域的交叉研究，专注于人工智能算法预测材料性能及实验验证。
    本文中负责撰写初稿和整理文献。
    YANG Yufan, Master’s student jointly educated by Qingdao University of Science and Technology and Institute of Process Engineering, Chinese Academy of Sciences. She is currently engaged in interdisciplinary research on ML and molecular simulation in the field of materials science, focusing on predicting material properties using artificial intelligence algorithms and validating them through experiments.
    In this paper, she is responsible for drafting the initial manuscript and collating the literatures. E-mail: yangyufan@ipe.ac.cn|李东祥，青岛科技大学化学与分子工程学院，博士，教授，应用化学系主任，应用化学专业负责人。主要研究方向为胶体与纳米界面结构、纳米粒子与聚合物纳米复合材料、等离激元共振纳米材料及其光学性能、功能纳米材料在新能源、环保和生物体系中的应用。
    本文中负责第3部分撰写和全文修改。
    LI Dongxiang, Ph.D., is a professor at the College of Chemistry and Molecular Engineering, Qingdao University of Science and Technology, Head of the Department of Applied Chemistry, and Director of the Applied Chemistry Program. His main research interests include colloid/interfacial nanostructures, nanoparticles and polymer nanocomposites, plasmonic nanomaterials and their optical properties, as well as the application of functional nanomaterials in new energy, environmental protection, and biological systems.
    In this paper, he is responsible for writing section 3 and revising the whole manuscript. E-mail: lidx@qust.edu.cn|周鹏，中国科学院过程工程研究所，博士，副研究员，专业领域涉及计算材料学、化学信息学、胶体界面化学，研究对象涵盖超分子组装纳米材料、玻璃态功能材料和生物医药材料。
    本文中负责文章整体架构设计与全文修改。
    ZHOU Peng, Ph.D., is an associate professor at the Institute of Process Engineering, Chinese Academy of Sciences. His research areas cover computational materials science, chemo-informatics, and colloid and interface chemistry, with research objects including supramolecular assembled nanomaterials, glassy functional materials, and biomedical materials.
    In this paper, he is responsible for designing the overall structure of the paper and revising the whole manuscript. E-mail: zhpayne@ipe.ac.cn
  • 基金资助:
    国家自然科学基金面上项目(22572195);东方之星青年科学家创新基金(DFZX202501)

Advances and Prospect of Computer-Aided Technology in Predicting the Properties of Glass Materials

YANG Yufan1,2(),LI Dongxiang1,*(),ZHOU Peng2,*()   

  1. 1 College of Chemistry and Molecular Engineering, Qingdao University of Science and Technology, Qingdao, Shandong 266042, China
    2 State Key Laboratory of Biopharmaceutical Preparation and Delivery, Institute of Process Engineering, Chinese Academy of Sciences, Beijing 100190, China
  • Received:2026-03-19 Online:2026-10-20 Published:2026-10-20

RichHTML

1

PDF

5

摘要：

【目的】为推动传统玻璃材料设计向智能化方向发展，为玻璃材料逆向设计提供参考，本文聚焦计算机技术在玻璃材料性能预测中的应用问题展开评述。【方法】以玻璃材料性能预测为主线，梳理传统机器学习、机器学习与分子模拟结合、玻璃材料逆向设计等领域研究现状，重点讨论机器学习建模、分子模拟辅助机理增强及预测模型可解释性分析的作用，对比不同方法优缺点并讨论发展趋势。【结果】研究表明，机器学习方法在玻璃性能预测中具备显著优势，但存在数据量少、来源分散且质量不均的问题；单纯基于成分的预测模型虽在部分性质上精度较高，但在结构机理表达、跨体系迁移和目标导向设计方面仍存在局限。【结论】为推进计算机辅助设计技术在玻璃材料领域的应用，应加强高质量数据库建设，引入更具物理约束的物理信息机器学习方法并结合深度学习知识迁移，推动从性能预测向目标导向的逆向设计发展。

关键词: 机器学习, 玻璃材料, 分子模拟, 逆向设计

Abstract:

[Purpose] To drive the intellectualization of traditional design and provide a reference for inverse design, this review focuses on the application of computer technologies in predicting the properties of glass materials. [Methods] Focusing on property prediction, we systematically summarize the research progress in traditional machine learning (ML), the fusion of ML with molecular simulation, and inverse design of glass materials. We highlight the roles of the traditional ML modeling, the simulation-aided mechanism enhancement and the interpretability analysis, compare their advantages and limitations, and discuss future development trends. [Results] ML exhibits significant advantages in property prediction. However, critical challenges persist, including scarce, scattered, and uneven-quality data. Furthermore, composition-only models, despite their high accuracy for certain properties, lack of capabilities in structural mechanism interpretation, cross-system transfer, and target-oriented design. [Conclusions] To advance computer-aided glass material research, we emphasize the need for high-quality databases. Integrating physics-informed ML (with physical constraints) and knowledge transfer in deep learning is crucial, which will facilitate the transition from property prediction to target-driven inverse design.

Key words: machine Learning, glass materials, molecular simulation, inverse design