For composite materials, the system response changes abruptly with a change in the properties of the material. Therefore, attaining significant knowledge about the effect of the material composition on the material properties is crucial. The researchers are looking for new computational methods which can predict these alterations so that the effort in experimental testing can be reduced. In this direction, this paper presents a robust and novel methodology of validating the estimation of the composite’s effective through a multi-scale approach by a set of standardized experimentation. These effective properties are estimated through the mean-field homogenization technique whose parameters are driven from the image analysis of Scanning Electron Microscopy (SEM) images. The predicted results are validated with the results obtained by the experimentation as per ASTM E1876 standard. The estimated error between the predicted properties and the experimental values increased with the increase of alumina particle fraction in the matrix. The mean-field homogenization lags behind the experiments for the parameters defined by the image analysis method than the experimental results. The upper bounds of the mean-field homogenization can be used for the composites with higher reinforcement volume fraction.

中文翻译：

---

复合材料平均场均质化有效性能的估计和实验验证

对于复合材料，系统响应会随着材料特性的变化而突然变化。因此，获得有关材料成分对材料特性影响的重要知识至关重要。研究人员正在寻找可以预测这些变化的新计算方法，从而减少实验测试的工作量。在这个方向上，本文提出了一种稳健而新颖的方法，通过一组标准化实验，通过多尺度方法来验证复合材料的有效性估计。这些有效特性是通过平均场均匀化技术估计的，其参数由扫描电子显微镜 (SEM) 图像的图像分析驱动。根据 ASTM E1876 标准，通过实验获得的结果验证了预测结果。预测性能与实验值之间的估计误差随着基体中氧化铝颗粒分数的增加而增加。平均场均匀化对于图像分析方法定义的参数比实验结果滞后。平均场均匀化的上限可用于具有较高增强体积分数的复合材料。平均场均匀化对于图像分析方法定义的参数比实验结果滞后。平均场均匀化的上限可用于具有较高增强体积分数的复合材料。平均场均匀化对于图像分析方法定义的参数比实验结果滞后。平均场均匀化的上限可用于具有较高增强体积分数的复合材料。