Analysis On The Reliable Degree Of Vertical Single Pile Bearing Capacity Based On The Random Field

In practice, parameters of soil property (including the elastic modulus, cohesion, and internal friction angle of the rock, soil, and weak dissection) have a significant spatial variability. They should be considered as the random field. Based on the random field characters of soil property, if the single random variables described are adopted, it may lead to underestimation of the reliability, and thus the design will be more conservative.That reliability analysis of pile foundation by the random field model overcomes the shortcomings of traditional analysis methods only by one safety factor to guarantee the reliability of the basis. It is more realistic to simulate the actual situation of the soil, with the correlation distance determined by random field model combined with CPT data. It succeeds in analyzing vertical pile baring capacity in conformity to the spatial characteristics of the soil. There is much practical value in the process of designation in optimizing design and improving the safety of buildings.This article analyses the characteristics of the recurrence space average methods in existence, and brings up the method to improve. Model the actual project by ANSYS finite element analysis software, and correlation distance calculated by CPT data, as well as the random field and the reliability theories. Sample calculating and analyzing factors affecting reliability index by Monte Carlo method, is more in line with the actual reliability index of the vertical pile bearing capacity. To get more referential value for pile design, here we compare the reliability index from empirical formula calculated by JC method and the one by the model using single random variables.The main conclusions are:1, Because the recurrence space average methods in existence often overlook the influence of sample intervals to the results of correlation distance calculated, this article put forward improvements by constantly adjusting sample intervals. When sample interval is close to the real correlation distance, the calculation results will be the correct correlation distance. At the same time the thickness of the soil meets the requirements of sample space,which is at least three times of the correlation distance. And the results are more credible.2, ANSYS reliability moduling is able to do well in the simulation of the reliability of the single pile in the soil with random field by the model using the random field. Force is the most influential factor of the reliability of the the single pile's bearing capacity in all parameters by the theory of random field, followed by the internal friction angle of pile tip soil. And in different soil side of the pile layers, the ultimate shaft friction and the internal friction angle of the thick layers have much impact.3, By comparing the result of the model using the random field with the one using random variables and the result of JC method, it gets such conclusions as follows: Model with the random field is more considerate of the soil parameters and the overall space, which is closer to the real situation. Influence of load and pile's own impact on the carrying capacity becomes prominent, especially higher relevant parameters in the load bearing capacity of all the implications. Force of soil parameters becomes more balanced, making a reliable higher project indicator and lower failure probability. This contributes to the reference for conservative project design. Neglect the random field of soil may underestimate the reliability, and the design lean to be conservative.