What question did this study set out to answer?

To assess how subsurface soil anisotropy and geometric arrangements influence lining forces in twin tunnels and develop machine learning models for early structural prediction.

March 18, 2020Underground SpaceOpen Access

Prediction of lining response for twin tunnels constructed in anisotropic clay using machine learning techniques

View Full Paper

Ask AI

Bookmark

To assess how subsurface soil anisotropy and geometric arrangements influence lining forces in twin tunnels and develop machine learning models for early structural prediction.
Simulated 682 twin-tunnel configurations via finite element analyses using the Norwegian Geotechnical Institute Active-Direct shear-Passive (NGI-ADP) anisotropic soil model.
Assessed five parameters: burial depth, tunnel center-to-center distance, undrained shear strength, stiffness ratio, and soil anisotropy ratio.
Trained multivariate adaptive regression splines and decision tree regressor algorithms on the simulation outputs to predict peak lining bending moments.
Center-to-center tunnel spacing and overlying burial depth served as the primary drivers governing lining bending moments and thrust forces, respectively.
Soil anisotropy significantly altered the induced structural forces, demonstrating that omitting anisotropic clay properties causes severe tunnel design errors.
Trained machine learning models successfully predicted maximum bending moments in the initial tunnel lining to support target project reliability standards.

Discussion

No takes yet. Share an insight, caveat, or question.

Implication

Finite element modeling study reveals key determinants of twin-tunnel lining forces in anisotropic clay, suggesting machine learning improves preliminary structural design.

Cite This Study

Zhang et al. (2020) studied this question.

synapsesocial.com/papers/6a82e9f58ddd99b9b523b37c https://doi.org/10.1016/j.undsp.2020.02.007

View Full Paper

Ask AI

Bookmark

Implication

Finite element modeling study reveals key determinants of twin-tunnel lining forces in anisotropic clay, suggesting machine learning improves preliminary structural design.

Prediction of lining response for twin tunnels constructed in anisotropic clay using machine learning techniques

Key Points

Discussion

Implication

Cite This Study

Prediction of lining response for twin tunnels constructed in anisotropic clay using machine learning techniques

Key Points

Discussion

Implication

Cite This Study

Also Consider

Also Consider