多相材料有效性能预测的高精度方法

AN APPROACH FOR PREDICTING THE EFFECTIVE PROPERTIES OF MULTIPHASE COMPOSITE WITH HIGH ACCURACY

有效介质方法是常用的细观力学方法之一.其可用于计算多相材料的有效性能,并建立材料微细观结构和宏观性能的定量关系;有助于指导新材料设计,减少试验工作量等.然而,当夹杂含量升高时,传统有效介质方法的计算精度下降.本文以两相材料为研究对象,提出一种新的参考介质,即:为更合理考虑不同夹杂颗粒间的相互作用,假定参考介质的应变是基体相平均应变和某一修正张量的双点积.在此基础上,推导了新参考介质下两相材料的有效模量表达式,并给出该修正张量的近似计算方法;通过反复更新参考介质,采用多层次均匀化思路,将本文方法进一步用于多相材料性能的预测.为验证方法的有效性,将预测结果与已有模型结果和试验数据进行对比.结果表明本文方法较已有方法更为合理、有效.当夹杂含量升高时,本文方法较传统有效介质方法的计算精度有所提升.

The efctive medium approach is one of the common micromechanical methods, which can be utilized to predict the material’s eective properties and set up the quantitative relationship between the material’s microstructures and macroscopic properties. It is helpful and meaningful for the new material design and reducing the (experimental)workload to use these micromechanical estimations of the material’s properties. However, the calculation accuracy will decline when the eective medium method is adopted to estimate the eective properties of the composite with high volume fraction of inclusions. Therefore, in this paper the two-phase composite is taken as the example firstly and the strain of the reference medium is assumed to be the product of the average strain of the matrix and a modifying tensor.Then the expressions of the efctive modulus are derived with the proposed reference medium. What’s more, the solutions for modifying tensor are reached by using the achievement we obtained recently. Further, through optimizing the reference medium repeatedly, with the help of multi-level homogenization scheme, the proposed modified method is extended to predict the properties of the multiphase composite with many types of inclusions. To verify our proposed framework,the predictions are compared with the experimental data and the results of existing models. The comparisons show that the estimations of the presented method are more reasonable and acceptable. When the volume fraction of inclusions is higher, the calculation accuracy of the presented method in this paper is better than those of the existing efctive medium methods.