Spatial Probabilistic Characteristics Of Geotechnical Parameters And Probability Analysis Of Consolidation Of Soft Clay Ground

The uncertainties in geotechnical engineering problems are so complex and particular that the application of reliability design theory in geotechnical engineering is difficult and some aspects need still to be further studied. The key to reliability analysis of geotechnical engineering is to model the soil profiles and to assess the uncertainties in geotechnical parameters, which has been one of the most leading directions in the development of geotechnical engineering. In this dissertation, a systematic study is carried out on the spatial probabilistic characteristics of geotechnical parameters and the probability analysis of consolidation of soft clay ground.It has been shown that based on the reliability check of experiment data, the optimum fit of probability model by finite contrast method and the estimation of distribution parameter by extended Bayesian method, the statistically optimal analysis of geotechnical parameter distribution can be realized and the collection of the district datum about probability distribution is the key point.The auto-correlation distance of soil parameter has the peculiarity of cross-isotropy. It is the most effective approach to calculate the vertical and the horizontal auto-correlation distance by using the static cone penetration test curve and the shear wave velocity respectively. The uncertainty involved in test methods of soil parameter can be effectively analyzed by correlation coefficient or random factor, the latter has a wider adaptability. On the basis of the engineering geological exploration datum, the district spatial probabilistic characteristics of soil parameters are established for Taiyuan silty clay and Pinghu muchy silty clay.The vertical and the horizontal coefficient of consolidation can be adequately modeled by either a lognormal distribution or a gamma distribution. A probability design method for one-dimensional consolidation of soft clay ground is proposed. According to random integral theory, the coefficient of consolidation is modeled as a homogeneous random field, and the spatial probability characteristics of consolidation of soft clay ground is studied. It shows that as far as one-dimensional consolidation is concerned, the influence of variation of auto-correlation distance on consolidation degree is insignificant. Generally, in the reliability analysis of consolidation of sand-drained ground, the influence of the vertical auto-correlation distance should beconsidered while the one of the horizontal auto-correlation distance may be ignored.It can be shown that the sensitivity of consolidation of sand-drained ground to the uncertainty of horizontal coefficient of consolidation is much stronger than to that of other soil parameters. An approximate probability method is then developed for consolidation of sand-drained ground in which both the effect of uncertainty and distribution type of horizontal coefficient of consolidation can be taken into consideration. The formulas for computing mean and deviation of the consolidation degree of sand-drained ground are also derived.Finally, it has been demonstrated that the reliability of consolidation of sand-drained ground can be analyzed by JC-method or Monte Carlo simulation method. A probability design procedure for sand-drained ground is established and the practical usage is illustrated by several examples.