岩石の破壊過程の計算モデルとその遅れ破壊への応用

組織構造の不均質性を考慮した破壊の確率過程論 (第1報)

抄録

Rock can be regarded as a kind of composite material which contains a sufficient number of structural particles with different mechanical properties. Accordingly the rock specimen was divided into 304 triangular elements and the Finite Element Method was used to simulate the failure process. Each element of model rock corresponds to a high or low elastic constant particle. Supposing that the failure process of each element is expressed as a 2-stage, i. e. 1-step Poisson process. a computer simulation of failure and deformation under a constant uniaxial tensile load was carried out by means of the theory of stochastic process and F. E. M.
M. Although each element has a most simple form of probability of survival, the strain-time behavior calculated by this computer simulation shows a similar tendency to that obtained experimentally in primary, steady-state, and tertiary creeps. The P-t diagram calculated, assuming that the model rock specimen is subjected to the eccentric load, shows good agreement to that experimentally obtained P-t diagram.
The P-t diagrams became concave to downward. This result can be explained as probability of survival being expressed by linear combination of two different 1-step-Poisson processes. However, it is apparent from this computer simulation that not only the above mentioned mechanism, but also some extraordinary distributions of stresses caused by inhomogeneity of rock specimen and an eccentric load make the P-t diagram concave to downward.