Composite materials inherently exhibit scatter in their basic characteristics such as the mechanical properties of constituent materials, fibers orientations, ply thicknesses, and applied loads. The uncertainties present in the basic parameters are available, in most cases, in the form of ranges or intervals. The use of the existing deterministic failure theories, which do not consider the observed variabilities in the input parameters, leads to poor failure assessment of composite materials. Several probabilistic failure models have been proposed in the past few decades. However, the probabilistic methods require a knowledge of the probability distributions of the basic variables which are not available in most practical systems. This work, for the first time, presents an interval-based failure analysis of composite materials using the truncation-based interval analysis and the universal gray system theory. These methods require only the lower and upper bounds of the uncertain parameters which are available in most practical systems. The application of the proposed interval-based failure models is demonstrated by considering two types of graphite/epoxy laminates: [0/±45/90] S and [0/90]2. This work shows that more realistic and meaningful failure assessment can be made using the universal gray system theory and the truncation-based interval analysis compared to the deterministic failure analysis.

中文翻译：

---

复合材料层压板的失效建模和分析：基于区间的方法

复合材料固有地表现出其基本特性的分散性，例如组成材料的机械性能、纤维取向、层厚度和施加的载荷。在大多数情况下，基本参数中存在的不确定性以范围或区间的形式存在。使用现有的确定性失效理论，不考虑输入参数中观察到的变量，导致复合材料的失效评估很差。在过去的几十年中，已经提出了几种概率故障模型。然而，概率方法需要了解基本变量的概率分布，而这在大多数实际系统中是不可用的。这项工作，第一次，介绍了使用基于截断的区间分析和通用灰色系统理论的复合材料基于区间的失效分析。这些方法只需要在大多数实际系统中可用的不确定参数的下限和上限。通过考虑两种类型的石墨/环氧树脂层压板：[0/±45/90] S 和 [0/90]2，证明了所提出的基于间隔的失效模型的应用。这项工作表明，与确定性故障分析相比，使用通用灰色系统理论和基于截断的区间分析可以进行更现实和有意义的故障评估。通过考虑两种类型的石墨/环氧树脂层压板：[0/±45/90] S 和 [0/90]2，证明了所提出的基于间隔的失效模型的应用。这项工作表明，与确定性故障分析相比，使用通用灰色系统理论和基于截断的区间分析可以进行更现实和有意义的故障评估。通过考虑两种类型的石墨/环氧树脂层压板：[0/±45/90] S 和 [0/90]2，证明了所提出的基于区间的失效模型的应用。这项工作表明，与确定性故障分析相比，使用通用灰色系统理论和基于截断的区间分析可以进行更现实和有意义的故障评估。