eva‎luation of bearing capacity of driven piles considering 3d spatial variability-A case study Abadan refinery tanks

9/18/2024 12:00:00 AM

Modeling the soil environment has always been one of the main challenges in geotechnical issues due to the limited information of field tests. The information obtained from geotechnical tests will also have uncertainty due to the unique physical and mechanical properties of the soil, the presence of water, the variety of different soil compositions, the complex behavior of stress-strain and other cases. One of the most common on-site tests to identify a construction is the cone penetration test (CPT). The main point of this method can be considered as providing continuous data, which can be used to correctly display and interpret the variability of soils in the Euclidean space with the data extracted from this experiment. Another main advantage of this test is the direct use of its data in estimating the bearing capacity of piles. According to this issue, in the case of three-dimensional modeling of a random field of the soil environment, the bearing capacity of the piles can be estimated at any desired point. In general, the random field is a method for generating random data, which random data is placed in space in a discrete or continuous manner. This theory is widely used for simulating and analyzing complex data. In this study, the bearing capacity of the pile at different points has been estimated by using the theory of random fields and using the cone penetration test data. The method used to generate this random field is based on the principles of the compactness of the cone penetration data (signal) and the Gaussian assumption of the field, which is based on the principles and theory of compressing sensing and separability in the autocorrelation function and assuming a The dense network is first calculated as the trend line, and then according to the information limitation, it is simulated using the Bayesian statistics and Markov Monte Carlo chain of the three-dimensional random field, and then the bearing capacity of the pile is calculated directly in the random field.

ظرفيت باربري شمع , ميدان تصادفي , تابع خودهمبستگي , يادگيري بيزي پراكنده , حسگري فشرده , زنجير ماركوف مونت كارلو , آزمون نفوذ مخروط

Pile bearing capacity , random field , autocorrelation function , sparse Bayesian learning , compact sensing , Monte Carlo Markov chain , , cone penetration test

Reza Gharibian

Mehran Karimpour Fard