Study On Bearing Capacity Of Deepwater Anchors Considering Spatial Variability Of Marine Clay

As the exploration and exploitation of oil and gas resources gradually develop into the deep sea,deep sea buoyant platforms have become the main choice instead of conventional fixed platforms.Deep sea buoyant platforms are generally fixed by connecting deepwater anchors in the seabed through a mooring system.Suction embedded plate anchors and torpedo anchors,as typical representatives of deepwater anchors,have the advantages of low cost,easy installation and large working depth,and have been widely used in anchoring of mooring systems.At present,the failure mechanism and bearing capacity of suction embedded plate anchors and torpedo anchors are mostly based on the assumption that marine clay is a deterministic soil.In fact,marine clay has obvious spatial variability due to various physical-chemical,environmental and geological effects.To this end,this paper combines the coupling Euler-Lagrangian finite element method,three-dimensional random field theory and Monte Carlo simulation,and uses the random finite element method to study the influence of the spatial variability of marine clay on the failure mechanism and bearing capacity of suction embedded plate anchors and torpedo anchors.The main research contents and results of this paper are as follows:The vertical pullout process of suction embedded plate anchors in spatially variable soils is simulated,and the failure mechanism of suction embedded plate anchors in the pullout process is studied.It is found that after considering the spatial variability of marine clay,there are three different characteristic flow mechanisms in the soil around suction embedded plate anchors.The influence of different spatial variability parameters on the vertical bearing capacity of suction embedded plate anchors was discussed,and it is found that the coefficient of variation and the vertical scales of fluctuation have a significant influence on the bearing capacity,while the horizontal scales of fluctuation has no obvious influence.Finally,from the perspective of bearing capacity,the failure probability under the safety factor is analyzed,and it is found that the safety factor in the existing standard cannot meet the needs of safety design.The horizontal pullout process of torpedo anchors in spatially variable soils is simulated.Based on the calculation framework of traditional Brom's method,Random Brom's method for calculating the horizontal bearing capacity of torpedo anchors considering the actual strength distribution in spatially variable soils is proposed.The model factor is introduced to evaluate the prediction performance of the RBM method,and the influence of different spatial variability parameters on the prediction performance of the RBM method is discussed.It is found that the prediction accuracy of the RBM method is mainly affected by two deviations: one is the deviation between the assumed failure mechanism and the actual failure mechanism;the other is the deviation between the sampling location and the actual installation location.Finally,based on the above two deviations,a calculation formula for the lower limit value of model factor is proposed for engineering practice.The multi-angle pullout of torpedo anchors in spatially variable soils is simulated,and the failure mechanism of torpedo anchors with different pullout angles is studied.It is found that the failure mechanism of torpedo anchors in spatially variable soils is similar to that in deterministic soils.When the pullout angle is increased from 0° to 90°,both of them show a transition from tilting failure to vertical failure.Based on the above simulation results,the failure envelope of torpedo anchors in spatially variable soils under three coefficients of variation is obtained,and probability statistics are carried out.Finally,a formula for the failure envelopes of torpedo anchors in spatially variable soils with different confidence levels is proposed,which provides a method for the reliability design of actual engineering.