Compared prediction of the experimental failure of a thin fibrous tissue by two macroscopic damage models

详细信息    查看全文

• 作者：A. Bel-Brunon ; M. Coret ; K. Bruy¨¨re-Garnier ; A. Combescure
• 关键词：Damage model ; Homogenization ; Fibrous tissue ; Numerical identification ; Stereo Digital Image Correlation ; Human liver capsule
• 刊名：Journal of the Mechanical Behavior of Biomedical Materials
• 出版年：2013
• 出版时间：November, 2013
• 年：2013
• 卷：27
• 期：Complete
• 页码：262-272
• 全文大小：1380 K

文摘

Several models for fibrous biological tissues have been proposed in the past, taking into account the fibrous microstructure through different homogenization methods. The aim of this paper is to compare theoretically and experimentally two existing homogenization methods - the Angular Integration method and the Generalized Structure Tensor method - by adapting them to a damage model for a planar fibrous tissue made of linear elastic and brittle fibers. The theoretical implementation of the homogenization methods reveals some differences once damage starts in the fibrous tissue; in particular, the anisotropy of the tissue evolves differently. The experimental aspect of this work consists in identifying the parameters of the damage model, with both homogenization methods, using inflation tests until rupture on a biological membrane. The numerical identification method is based on the simulation of the tests with the real geometry of the samples and the real boundary conditions computed by Stereo Digital Image Correlation. The identification method is applied to human liver capsule. The collagen fibers Young's modulus (19¡À6 MPa) as well as their ultimate longitudinal strain (33¡À4 % ) are determined; no significant difference was observed between the two methods. However, by using the experimental boundary conditions, we could observe that the damage progression is faster for the Angular Integration version of the model.