期刊
COMPOSITE STRUCTURES
卷 278, 期 -, 页码 -出版社
ELSEVIER SCI LTD
DOI: 10.1016/j.compstruct.2021.114683
关键词
Uncertainty quantification; Composite; Interlaminar fracture; Mode-I; DCB; Mode-II
This paper presents a comprehensive stochastic analysis of pure and mixed mode interlaminar fracture of fibre-reinforced polymer (FRP) composites using different uncertainty quantification approaches. The study aims to evaluate the accuracy and computational cost of stochastic reduced order models in finite element modelling of FRP composite fracture. The results show that combining a stochastic reduced order model with a surrogate model can achieve computational efficiency with a high level of accuracy even with a limited number of samples.
A comprehensive stochastic analysis of pure and mixed mode interlaminar fracture of fibre-reinforced polymer (FRP) composites, using different uncertainty quantification approaches, is presented. The primary aim of this work is to evaluate the accuracy and computational cost of stochastic reduced order models in the finite element modelling of FRP composite fracture. The Monte Carlo method, with different sampling methods was considered as a reference method for comparison purposes. By comparing the descriptive statistics of uncertain quantities of interest, the advantages and drawbacks of these methods is revealed. Double Cantilever Beam (DCB), End Notch Flexure (ENF) and Mixed Mode Bending (MMB) tests were simulated using a stochastic cohesive zone model. Fundamental characteristics of the cohesive zone model, such as the interlaminar fracture toughness and cohesive strength, were assumed as uncertain sources, and crack extension and critical force were considered as uncertain quantities of interest. The use of a stochastic reduced order model combined with a surrogate model is shown to be computationally efficient, for a given level of accuracy, even with a limited number of samples.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据