A mechanical particle model for analyzing rapid deformations and fracture in 3D fiber materials with ability to handle length effects
2014 (English)In: International Journal of Solids and Structures, ISSN 0020-7683, E-ISSN 1879-2146, Vol. 51, no 11-12, 2244-2251 p.Article in journal (Refereed) Published
A mechanical model for analyses of rapid deformation and fracture in three-dimensional fiber materials is derived. Large deformations and fractures are handled in a computationally efficient and robust way. The model is truly dynamic and computational time and memory demand scales linearly to the number of structural components, which make the model well suited for parallel computing. The specific advantages, compared to traditional continuous grid-based methods, are summarized as: (1) Nucleated cracks have no idealized continuous surfaces. (2) Specific macroscopic crack growth or path criteria are not needed. (3) The model explicitly considers failure processes at fiber scale and the influence on structural integrity is seamlessly considered. (4) No time consuming adaptive re-meshing is needed. The model is applied to simulate and analyze crack growth in random fiber networks with varying density of fibers. The results obtained in fracture zone analyses show that for sufficiently sparse networks, it is not possible to make predictions based on continuous material assumptions on a macroscopic scale. The limit lies near the connectivity l(c)/L = 0.1, where is the ratio between the average fiber segment length and the total fiber length. At ratios l(c)/L < 0.1 the network become denser and at the limit l(c)/L -> 0, a continuous continuum is approached on the macroscopic level.
Place, publisher, year, edition, pages
2014. Vol. 51, no 11-12, 2244-2251 p.
Dynamic high speed fracture, Random fiber network, Large deformations
Research subject Engineering science with specialization in Applied Mechanics
IdentifiersURN: urn:nbn:se:uu:diva-225505DOI: 10.1016/j.ijsolstr.2014.02.031ISI: 000335272100019OAI: oai:DiVA.org:uu-225505DiVA: diva2:728539
FunderSwedish Research Council, 2010-4348