Radu Popescu -
Jean H. Prevost -
George Deodatis
Department of Civil Engineering and Operations Research
Princeton University, Princeton, New Jersey 08544
One of the major problems in geotechnical engineering is estimating the geomechanical properties of soil materials, for use in modeling. Aside from the measurement induced bias and even within assumed homogeneous soil strata, natural soils are variable in their properties. The scarcity in field test results, characteristic to geotechnical engineering, and a sufficiently large degree of disorder exhibited by soil properties leads to the use of statistical methods in describing the distribution of those properties within a ``stochastically homogeneous'' soil zone.
The study mainly deals with the stochastic variability of soil properties, but also addresses other uncertainties affecting the geotechnical design. Methods for (1) field data analysis for estimating the probabilistic characteristics of spatial variability, and (2) digital simulation of non-Gaussian random fields representing the stochastic distribution of various soil properties are proposed.
Monte Carlo simulations of soil liquefaction are performed for a saturated soil deposit subjected to seismic excitation, and the predicted excess pore pressures are compared with similar results obtained from deterministic analyses. It is concluded that a more realistic pattern of soil liquefaction occurrence, and higher pore water pressure build-up are predicted if the stochastic variability of soil properties is accounted for.
| Back to: |
|
|