Undrained capacity of skirted circular foundations under fully three-dimensional loading

Chen, Hongzhen; Shen, Zuowei; Wang, Le; Tian, Yinghui; Chu, Xuanxuan

doi:10.1016/j.compgeo.2023.105261

Cited by 8 publications

(1 citation statement)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The random finite element method (RFEM) and random finite difference method (RFDM) have been widely used in the probabilistic analysis of geotechnical engineering, considering soil spatial variability. These methods have been applied in slope stability [10][11][12][13], foundation bearing capacity [14][15][16][17][18], tunnel stability and deformation [19][20][21], seismic response of structures [22][23][24], etc. The non-stationary characteristic exists in seabed soil parameters, as shown by Hossain and Randolph [25].…”

Section: Introductionmentioning

confidence: 99%

Prediction of Undrained Bearing Capacity of Skirted Foundation in Spatially Variable Soils Based on Convolutional Neural Network

et al. 2023

View full text Add to dashboard Cite

Skirted foundations are widely used in offshore and subsea engineering. Previous studies have shown that soil undrained shear strength variability has a notable impact on probabilistic analyses of skirted foundation bearing capacity. This study proposes an efficient machine-learning method to predict the uniaxial bearing capacity factors of skirted foundations under pure horizontal and moment loads, without relying on traditional time-consuming random finite element methods. A two-dimensional convolutional neural network is adopted to capture the potential correlation between soil random fields and bearing capacity factors. The proposed CNN-based model exhibits satisfactory prediction performance with regard to coefficients of variation and scale of fluctuations in two directions. Specifically, coefficient of determination (R2) values exceed 0.97, while root mean square error (RMSE) values remain below 0.13 for the surrogate model. In addition, more than 96% of the predictions are associated with a relative error of 5% or less, providing evidence of the proposed 2D-CNN model’s satisfactory prediction performance.

show abstract

Section: Introductionmentioning

confidence: 99%

Prediction of Undrained Bearing Capacity of Skirted Foundation in Spatially Variable Soils Based on Convolutional Neural Network

et al. 2023

View full text Add to dashboard Cite

show abstract

Interaction domains for capacity- and performance-based design of pile groups

Cesaro,

Di Laora,

Iodice

et al. 2024

Acta Geotech.

View full text Add to dashboard Cite

The aim of the paper is to propose design tools for pile groups when vertical and moment loads are applied simultaneously. This is done by presenting a method to derive the exact solution for the failure locus in the generalized force space, including a vertical load and a moment with components along two axes, to employ for capacity requirements which helps to get rid of overconservative assumptions usually adopted in design. Moreover, closed-form solutions for performance-based interaction domains are also provided allowing to detect all the combinations of external loads associated with a given value of settlement and/or rotation of the foundation that is extremely useful when the design is ruled by serviceability requirements. The application of these straightforward formulae provides results which are in very good agreement with the more rigorous solutions derived from a numerical procedure developed ad hoc for the problem at hand.

show abstract