双随机分布细观分析模型与复合材料性能预报

发展了一种细观力学有限元分析方法——拟真实的参数化双随机分布模型, 该模型综合考虑了纤维增强树脂基复合材料的真实微结构特点和纤维单丝综合力学性能测试结果的离散性特征, 模拟了复合材料中纤维排列和强度分布的随机性。借助移动窗口法研究了该参数化双随机分布模型的可靠性, 确定了其代表性体积单元的尺寸。基于能量法原理推导了单向复合材料的弹性模量预测公式, 结合能量法和渐进失效分析方法, 利用该细观力学有限元方法分别预测了单向纤维增强树脂基复合材料T300/5228的弹性模量和强度性能。数值模拟结果和大部分试验结果吻合良好, 表明发展的细观力学有限元方法能够较好地预测复合材料的力学性能。 

Double-random distribution model and properties prediction of composites

A micromechanical finite element method——parameterized double-random distribution model of virtual reality was developed. Integrated with the real microstructure character of fiber reinforced resin matrix composites and discrete characteristics of testing results about mechanical properties for monofilaments, random distribution of fiber arrangement and fiber strength were both simulated in the model. The reliability of parameter random distribution model was studied with moving window method and the size of its representative volume element was determined. The predicted formulas of elastic modulus for unidirectional composites were derived based on energy method. Combined with energy method and progressive failure analysis, elastic modulus and strength properties of unidirectional fiber reinforced resin matrix composite T300/5228 were respectively predicted by the micromechanical finite element method. The numerical results agree well with most experimental data, which shows that the micromechanical finite element method can predict the mechanical properties of composites correctly.