Scholarlink

<p>This paper investigates the mechanical behavior of fiber-sand mixtures (FSM) by comprehensively examining the effects of fiber volumetric content (FC) and fiber length (FL) from both macroscopic and microscopic perspectives using the discrete element method (DEM). A hybrid digital photogrammetry and deep learning approach integrating YOLOv5 and U-Net is developed to rapidly identify particle contours and construct realistic shape databases for fibers and sand grains. Based on these databases, numerical biaxial compression tests are conducted on FSM samples with different FC and FL. The results show that increasing FC and FL significantly enhances the peak and residual stress ratios, leading to increases in peak and residual friction angles of up to 8.5% and 6.2%, respectively. Higher FC values also promote volumetric shrinkage, while longer fibers delay volume contraction by providing additional bridging and rotational restraint. Microscopic analysis reveals that increasing FC reduces sand-sand coordination numbers and increases sand-fiber contacts, while longer fibers contribute to a higher sliding contact ratio and improve internal stability by restricting particle rearrangement. The findings elucidate the micro-mechanisms governing strength improvement and deformation behavior in FSM and provide guidance for optimizing fiber reinforcement strategies in geotechnical engineering applications.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

DEM study on the effect of fiber length and content on the macro-micro mechanical behavior of fiber-sand mixture

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年1月13日

查看全文
  • Fengling Tan,
  • Guangjing Yin,
  • Pengpeng Zhao,
  • Guocheng Sun,
  • Zhe Li,
  • Zongtang Zhang

摘要

This paper investigates the mechanical behavior of fiber-sand mixtures (FSM) by comprehensively examining the effects of fiber volumetric content (FC) and fiber length (FL) from both macroscopic and microscopic perspectives using the discrete element method (DEM). A hybrid digital photogrammetry and deep learning approach integrating YOLOv5 and U-Net is developed to rapidly identify particle contours and construct realistic shape databases for fibers and sand grains. Based on these databases, numerical biaxial compression tests are conducted on FSM samples with different FC and FL. The results show that increasing FC and FL significantly enhances the peak and residual stress ratios, leading to increases in peak and residual friction angles of up to 8.5% and 6.2%, respectively. Higher FC values also promote volumetric shrinkage, while longer fibers delay volume contraction by providing additional bridging and rotational restraint. Microscopic analysis reveals that increasing FC reduces sand-sand coordination numbers and increases sand-fiber contacts, while longer fibers contribute to a higher sliding contact ratio and improve internal stability by restricting particle rearrangement. The findings elucidate the micro-mechanisms governing strength improvement and deformation behavior in FSM and provide guidance for optimizing fiber reinforcement strategies in geotechnical engineering applications.