欢迎访问《金属矿山》杂志官方网站,今天是 分享到:
×

扫码分享

金属矿山 ›› 2026, Vol. 55 ›› Issue (4): 93-107.

• 采矿工程 • 上一篇    下一篇

基于机器学习的3D 打印透明类岩石材料力学特性模拟及制备优化#br#

何艺林1 钱自卫1 谢世平2 刘飓风1   

  1. 1. 中国矿业大学资源与地球科学学院,江苏 徐州 221008;2. 长沙矿山研究院有限责任公司,湖南 长沙 410000
  • 出版日期:2026-04-15 发布日期:2026-05-08
  • 通讯作者: 钱自卫(1986—),男,副教授,博士,硕士研究生导师。
  • 作者简介:何艺林(2002—),男,硕士研究生。
  • 基金资助:
    国家自然科学基金资助项目(编号:42172283)。

Mechanical Properties Simulation and Preparation Optimization of 3D Printing Transparent Rock-like Materials Based on Machine Learning#br#

HE Yilin1 QIAN Ziwei1 XIE Shiping2 LIU Jufeng1   

  1. 1. School of Resources and Geosciences,China University of Mining and Technology,Xuzhou 221008,China;
    2. Changsha Institute of Mining Research Co. ,Ltd. ,Changsha 410000,China
  • Online:2026-04-15 Published:2026-05-08

摘要: 本研究针对岩石物理模拟材料透明度与力学性能调控的矛盾,构建基于光固化树脂的透明3D 打印类
岩石材料体系。基于3D 打印树脂材料制作标准试样,通过开展双因素五水平共25 组单轴压缩试验,获得了材料的
透光率、弹性模量、抗压强度以及峰值应变等数据,探讨了不同材料组分对这些性能指标的敏感性。随后采用随机森
林算法与NSGA-Ⅱ多目标优化方法,系统探究了柔性树脂与清洗剂配比对材料光学—力学性能的耦合机制。研究发
现,柔性树脂含量对透光率与峰值应变起主导调控作用;清洗剂主要影响了弹性模量与抗压强度的衰减规律。多目
标优化结果揭示了“高透光率—高弹性模量”与“高峰值应变—高抗压强度”的性能权衡规律,筛选出与砂岩、泥岩、煤
岩力学特征匹配的最优配方,实现了透光率与力学参数的协同调控,为岩体变形破坏的可视化研究奠定了材料基础。

关键词: 类岩石材料 , 相似材料 , 机器学习 , 3D 打印 , 力学特性 , 材料制备

Abstract: In this study,a transparent 3D printing rock-like material system based on light-cured resin was constructed to
solve the contradiction between the transparency and mechanical properties of rock physical simulation materials. Based on the
3D printing resin material,a standard sample was made. Through 25 sets of uniaxial compression tests with two factors and five
levels,the data of transmittance,elastic modulus,compressive strength and peak strain of the material were obtained,and the
sensitivity of different material components to these performance indexes was discussed. Subsequently,the random forest algorithm
and NSGA-Ⅱ multi-objective optimization method were used to systematically explore the coupling mechanism of the ratio
of flexible resin and cleaning agent on the optical-mechanical properties of materials. It is found that the content of flexible
resin plays a dominant role in the regulation of light transmittance and peak strain. The cleaning agent mainly affects the attenuation
law of elastic modulus and compressive strength. The multi-objective optimization results reveal the performance trade-off
law of ′high transmittance-high elastic modulus′ and ′high peak strain-high compressive strength′,and screen out the optimal
formula matching the mechanical characteristics of sandstone,mudstone and coal rock. The synergistic regulation of transmittance
and mechanical parameters is realized,which lays a material foundation for the visualization study of rock mass deformation
and failure.

Key words: rock-like materials,similar material,machine learning,3D printing,mecharical preperty,material perparation

中图分类号: