Symbolic regression based prediction of anisotropic closure in deep tunnels - Archive ouverte HAL
Article Dans Une Revue Computers and Geotechnics Année : 2024

Symbolic regression based prediction of anisotropic closure in deep tunnels

Lina-María Guayacán-Carrillo
Jean Sulem
  • Fonction : Auteur
  • PersonId : 1148682

Résumé

This work investigates the applicability of Genetic Programming with Symbolic Regression to analyse the closure evolution of tunnels during and after excavation. Special attention is given to anisotropic closure evolution, which depends on the anisotropy of the initial stress state and the intrinsic anisotropy of the rock mass formation. A methodology is proposed that takes into account the information recorded during excavation to train the algorithm and find a free-form simple mathematical expression that captures the closure evolution over time. The proposed methodology is applied to two case studies of deep tunnels with high anisotropic convergence evolution. The results obtained show that this approach performs well with the small dataset used in this work. The proposed approach is an interesting tool to improve the understanding and prediction of the ground response.

Dates et versions

hal-04560768 , version 1 (26-04-2024)

Identifiants

Citer

Lina-María Guayacán-Carrillo, Jean Sulem. Symbolic regression based prediction of anisotropic closure in deep tunnels. Computers and Geotechnics, 2024, 171, pp.106355. ⟨10.1016/j.compgeo.2024.106355⟩. ⟨hal-04560768⟩
30 Consultations
0 Téléchargements

Altmetric

Partager

More