Scholarlink

<p>In this study, three data-driven models, artificial neural network, radial basis function (RBF) model and Kriging model, are developed to achieve accurate and fast prediction of the stress–strain behavior of ceramic matrix composite (CMC). These models are trained based on finite element analysis to generate sample stress–strain curves. The inputs of the data-driven models are mechanical properties of fibers and matrix, parameters of yarn structure and defects, and the outputs are stress–strain curves under specific loading conditions. Comparisons between the data-driven models indicate that RBF performs best in predicting stress–strain curves for both longitudinal stretch and in-plane shear. The global sensitivity analysis shows that the modulus and shear strength of the matrix and the modulus and tensile strength of the fibers have a significant effect on the mechanical properties of CMC. The geometrical parameters of the yarn affect the mechanical properties of CMC by influencing the volume fraction of fibers, but increasing the yarn width or decreasing the thickness also enhances the tensile properties of CMC when the volume fraction is constant. The mechanical properties of CMC are affected by both the porosity and the size of the defects.</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

Data-driven prediction of stress–strain behavior in defect-containing Cf/SiBCN ceramic matrix composites: accounting for multi-factor influences

  • 期刊论文

  • 发表日期：

    2025年11月19日

查看全文
  • Yuncan Pan,
  • Xin Liu,
  • Wei Ye,
  • Enze Jin,
  • Jianqiang Xin,
  • Jianyao Yao

摘要

In this study, three data-driven models, artificial neural network, radial basis function (RBF) model and Kriging model, are developed to achieve accurate and fast prediction of the stress–strain behavior of ceramic matrix composite (CMC). These models are trained based on finite element analysis to generate sample stress–strain curves. The inputs of the data-driven models are mechanical properties of fibers and matrix, parameters of yarn structure and defects, and the outputs are stress–strain curves under specific loading conditions. Comparisons between the data-driven models indicate that RBF performs best in predicting stress–strain curves for both longitudinal stretch and in-plane shear. The global sensitivity analysis shows that the modulus and shear strength of the matrix and the modulus and tensile strength of the fibers have a significant effect on the mechanical properties of CMC. The geometrical parameters of the yarn affect the mechanical properties of CMC by influencing the volume fraction of fibers, but increasing the yarn width or decreasing the thickness also enhances the tensile properties of CMC when the volume fraction is constant. The mechanical properties of CMC are affected by both the porosity and the size of the defects.