COMPUTING QUASIBRITTLE FAILURE PROBABILITY: FROM NANO TO MACRO
In civil, aeronautical and naval engineering, in protection from natural hazards, and in micro-electronics and MEMS, and one must ensure an extremely small failure probability, < 10-6. How to do that is adequately understood only for the limiting special cases of perfectly brittle or ductile beha...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Text |
| Language: | English |
| Published: |
2008
|
| Subjects: | |
| Online Access: | http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.180.2279 http://www.iacm-eccomascongress2008.org/admin/files/fileabstract/a3542.pdf |
| Summary: | In civil, aeronautical and naval engineering, in protection from natural hazards, and in micro-electronics and MEMS, and one must ensure an extremely small failure probability, < 10-6. How to do that is adequately understood only for the limiting special cases of perfectly brittle or ductile behaviors, for which the structural strength distribution is necessarily Weibullian or Gaussian, respectively. Presented is a computational approach to do that for the broad class of quasibrittle structures, which have brittle constituents with material inhomogeneities of non-negligible size, and include structures made of concrete (as the archetypical case), rocks, fiber composites, wood, toughened ceramics, rigid foams, sea ice, stiff soils, snow slabs, etc., as well as metals and ceramics on approach to nano-scale. It is shown that, for such structures, the strength distribution is transitional, having a Weibull asymptote (with zero threshold) on the left and a Gaussian (normal) asymptote on the right, with the transition center shifting from left to right as a function of structure size and geometry. The consequence is that, for quasibrittle structures, the understrength partial safety factor must also be |
|---|