Closed-form seismic earth pressure solutions via adaptive limit analysis and hybrid learning models
Article
| Article Title | Closed-form seismic earth pressure solutions via adaptive limit analysis and hybrid learning models |
|---|---|
| ERA Journal ID | 40357 |
| Article Category | Article |
| Authors | Nguyen, Tan, Shiau, Jim and Bui-Ngoc, Tram |
| Journal Title | Acta Geotechnica |
| Number of Pages | 25 |
| Year | 2025 |
| Publisher | Springer |
| Place of Publication | Germany |
| ISSN | 1861-1125 |
| 1861-1133 | |
| Digital Object Identifier (DOI) | https://doi.org/10.1007/s11440-025-02858-7 |
| Web Address (URL) | https://link.springer.com/article/10.1007/s11440-025-02858-7 |
| Abstract | The need for accurate seismic earth pressure solutions in earthquake geotechnical engineering demands a cost-effective method. This paper employs a three-stability-factor approach to determine seismic earth pressures by considering cohesion, surcharge, and unit weight effects that is analogous to Terzaghi’s traditional superposition method for bearing capacity determination proposed by Terzaghi. To achieve this goal, adaptive finite element limit analysis is used to explore seismic earth pressure intricacies using both upper-bound and lower-bound approaches. Numerical findings highlight the influence of internal friction angle, wall roughness, and surcharge pressure on seismic earth pressure factors. Distinct failure mechanisms of smooth and rough retaining walls under seismic loads offer vital insights for practical design. Incorporating cutting-edge machine learning techniques such as Bayesian regularization feed forward neural network and multivariate adaptive regression splines, a series of closed-form solutions for stability factors is established. These data-driven solutions ensure precision, simplicity, and efficiency in determining seismic lateral earth pressure. This approach transcends theoretical boundaries, providing insights for designing stable retaining walls in seismic zones. Rigorous validation against published results confirms the accuracy and reliability of the developed solutions. This research represents a significant advancement in seismic design methodologies, contributing to enhanced infrastructure resilience in the face of seismic challenges. |
| Keywords | Adaptive finite element limit analysis ; Closed-form solution; Machine learning; Retaining wall design; Seismic earth pressures; Stability factors |
| Contains Sensitive Content | Does not contain sensitive content |
| ANZSRC Field of Research 2020 | 400502. Civil geotechnical engineering |
| Public Notes | Files associated with this item cannot be displayed due to copyright restrictions. |
| Byline Affiliations | Ton Duc Thang University, Vietnam |
| School of Science, Engineering and Digital Technologies - Engineering | |
| Van Lang University, Viet Nam |
https://research.usq.edu.au/item/100w63/closed-form-seismic-earth-pressure-solutions-via-adaptive-limit-analysis-and-hybrid-learning-models
52
total views0
total downloads9
views this month0
downloads this month