Machine learning approaches for stability prediction of rectangular tunnels in natural clays based on MLP and RBF neural networks

Intelligent Systems with Applications, Journal Year: 2024, Volume and Issue: 21, P. 200329 - 200329

Published: Jan. 13, 2024

In underground space technology, the issue of tunnel stability is a fundamental concern that significantly causes catastrophe. Owing to sedimentation and deposition processes, strengths clays are anisotropic, where magnitudes undrained shear in vertical horizontal directions different. The anisotropic (AUS) model effective at considering anisotropy clayey soils when analyzing geotechnical issues. This study aims assess rectangular tunnels by adjusting dimensionless overburden factor, cover-depth ratio, width-depth ratio clay with various strength ratios. analysis these involves employing finite element limit AUS identify planes soil collapse response aforementioned variations. addition, this presents development soft-computing models utilizing artificial neural networks (ANNs) forecast across combinations input parameters. findings presented form design charts, tables, facilitate practical applications.

Language: Английский

---

Bearing capacity prediction of open caissons in anisotropic clays utilizing a deep neural network coupled with a population based training approach

Results in Engineering, Journal Year: 2025, Volume and Issue: unknown, P. 104323 - 104323

Published: Feb. 1, 2025

Language: Английский

---

Stability evaluation of elliptical tunnels in natural clays by integrating FELA and ANN

Results in Engineering, Journal Year: 2023, Volume and Issue: 19, P. 101280 - 101280

Published: July 4, 2023

The stability of tunnels in clayey soil is a major concern for underground space technology. Clay has anisotropy shear strength induced by depositional and sedimentation processes. For the numerical analysis geotechnical problems, anisotropic undrained (AUS) model can account this soils. In study, elliptical tunnel (stability factor: σs-σt/suc) with varying shape (width-depth ratio: B/D) placed at different embedment depths (cover-depth C/D) clay (anisotropic re) dimensionless overburden factor (overburden γD/suc) evaluated using finite element limit AUS model. failure planes are also above variations. Based on outcome, artificial neural network (ANN) utilized to establish equation predicting shapes (i.e., width-depth ratio), overburden, cover-depth ratio, ratio clay. present study results presented as design charts, tables, equations so that they be used practice.

Language: Английский

---

Hybrid artificial neural network models for bearing capacity evaluation of a strip footing on sand based on Bolton failure criterion

Transportation Geotechnics, Journal Year: 2024, Volume and Issue: 48, P. 101347 - 101347

Published: Aug. 23, 2024

Language: Английский

---

Data-driven modelling of bearing capacity of footings on spatially random anisotropic clays using ANN and Monte Carlo simulations

International Journal of Geotechnical Engineering, Journal Year: 2024, Volume and Issue: unknown, P. 1 - 17

Published: June 11, 2024

The fundamental issue of bearing capacity footings on anisotropic clays holds significant importance in geotechnical engineering. Previous investigations predominantly focused deterministic analyses, disregarding the spatial variability soil. A probabilistic analysis is conducted this paper, incorporating clays. To achieve this, Random Adaptive Finite Element Limit Analysis (RAFELA) and Monte Carlo simulations are utilised to capture full spectrum potential outcomes under parametric uncertainty. impact soil strength explored across three input parameters such as ratios, coefficients variation, dimensionless correlation lengths. In order establish surrogate models capable predicting random clays, Artificial Neural Network (ANN) developed. use proposed ANN presents a more convenient computationally efficient approach for ultimate vertical load spatially

Language: Английский

---

Application of artificial neural networks for predicting lateral and uplift capacity of buried rectangular box carrying pipelines

Marine Georesources and Geotechnology, Journal Year: 2024, Volume and Issue: 42(12), P. 1907 - 1923

Published: Jan. 5, 2024

Urban and offshore subterranean comprehensive pipelines are currently utilized frequently as infrastructure for fitting different engineering used electricity, signal transmission, natural gas, petroleum, heating, water supply, drainage systems. This study utilizes the finite element limit analysis (FELA) artificial neural network (ANN) to evaluate pull-out capacity factor of buried rectangular box carrying with internal inclined force in cohesive soil. In FELA, rigorous upper bound lower solutions performed achieve exact result. this study, a dimensionless parametric is carried out by considering effect five parameters viz. depth ratio, width-depth overburden factor, inclination angles, interface on box, which can be called pipeline. Two significant samples investigated selected soil failure pattern (failure envelope) using shear dissipation pipeline subjected loading. Using FELA results, predictive equations proposed ANN, then sensitivity examine each essential parameter stability underground The ANN model pipe loading presented design charts practice further investigation.

Language: Английский

---

An ANN-based model for constructing undrained failure envelopes for circular piles under general loading

Ships and Offshore Structures, Journal Year: 2025, Volume and Issue: unknown, P. 1 - 23

Published: Jan. 30, 2025

Language: Английский

---

Coupled Finite Element Analysis and Multivariate Adaptive Regression Splines for predicting the bearing capacity of conical footings on slopes

Journal of Physics Conference Series, Journal Year: 2025, Volume and Issue: 2949(1), P. 012025 - 012025

Published: Feb. 1, 2025

Abstract Exploring the behavior of conical footings is a significant aspect geotechnical engineering, particularly in supporting wind turbine towers on mountain slopes. This study employs Finite Element Analysis (FEA) within Plaxis 3D to investigate footings. The Mohr-Coulomb material assumed, and research focuses two pivotal parameters soil shear strength: c (cohesion) φ (friction angle). aims assess impacts geometry parameters, consisting angle ( α ), setback ratio b/B slope β failure mechanisms footing. Furthermore, Multivariate Adaptive Regression Splines (MARS) models are implemented with FEA dataset propose predictive formula UBC (ultimate bearing capacity) factor, indicating relationship between these five input outcome UBC. A sensitivity analysis was also examined, revealing contribution each variable

Language: Английский

---

Soft Computing-Based Models for Estimating the Ultimate Bearing Capacity of an Annular Footing on Hoek–Brown Material

Arabian Journal for Science and Engineering, Journal Year: 2023, Volume and Issue: 49(4), P. 5989 - 6006

Published: Dec. 28, 2023

Language: Английский

---

Soft Computing-Based Models for Estimating Undrained Bearing Capacity Factor of Open Caisson in Heterogeneous Clay

Geotechnical and Geological Engineering, Journal Year: 2024, Volume and Issue: 42(6), P. 5335 - 5361

Published: May 6, 2024

Language: Английский

---