基于有限元计算细观力学的复合材料宏观性能的一体化预测

基于有限元计算细观力学理论,以MSC.PATRAN为平台,利用PCL实现了RVE的参数化自动建模,建立了RVE库。并根据不同的细观力学方法,通过在后台对RVE边界条件的设置及有限元计算,集成了相应的计算程序,从而实现了复合材料宏观性能的一体化预测。最后,以单向复合材料的宏观弹性模量预测为例,展示了一体化预测系统。

A Better Integrated Prediction of Macroscopic Properties of Fiber Reinforced Composites Based on Finite Element Computational Micromechanics (FECM)

Aim.Integrated predictions in the past are,in our opinion,not quite satisfactory.We now present what we believe to be a better one. In the full paper,we explain our better one in some detail.In this abstract,we just add some pertinent remarks to naming the first three sections of the full paper.The first section is: the computational method of FECM.In this section,we present the relationships among the values of micro-local field and those of macro average field.The second section is: the parameterized automatic modeling of a representative volume element(RVE).Its three subsections are: establishing the finite element model of the RVE(subsection 2.1),the inputting of RVE parameters(subsection 2.2) and establishing the RVE library(subsection 2.3).In subsection 2.1,the parameterized finite element modeling of RVEs is realized by the Boolean operations as shown in Fig.1 in the full paper and then integrated into the RVE library in PATRAN environment using PATRAN command language (PCL).In subsection 2.3,Fig.2 shows the four different-pattern RVEs of fiber-reinforced composites now available in our RVE library.The third section is: our integrated prediction of macroscopic properties.In section 3,Fig.3 shows the schematic of the FORTRAN pre-processing and the C post-processing.Finally,we give as a numerical example the macroscopic elastic moduli prediction of unidirectional fiber-reinforced composites and compare the numerical values with the experimental values.The numerical results,given in Table 2,show preliminarily that our integrated prediction method is quite satisfactory for engineering application.