Using Artificial Neural Networks to Predict Seismic Shear-Induced Pore Water Pressure
Using Artificial Neural Networks to Predict Seismic Shear-Induced Pore Water Pressure
The very loose sandy subsoil is observed in the UAE region. The high magnitude of the undrained cyclic shear stresses produced by the earthquake or explosion is a reason for liquefying the confined, low-permeable, saturated, noncohesive soil; this phenomenon is known as liquefaction. With minimizing liquefaction resistance and increasing the liquefaction zone of the sandy subsoil model, pore water pressure is raised, the sand particle contact is reduced, shear strength is immediately reduced, etc. This paper aims to predict the fluctuation of pore water pressure of the loose saturated sand by using the Artificial Neural Network (ANN) and advanced mathematical methods. The input for the ANN modeling was designed using the experimental data published in the literature. To establish the ANN model, acceleration history (g) is applied at a selected point, shear stress is applied at a designated point, and dynamic response of the model is used as input. The results indicate that using a suitable advanced method is an efficient technique to be used for predicting the fluctuation of the pore water pressure with minimized error.
Abdoullah Namdar; Omer Mughieda
4th Asia-Pacific Conference on Physical Modelling in Geotechnics (ACPMG2024)
Other
https://doi.org/10.53243/ACPMG2024-27
Contact Information
- Secretary General - Dr A M McNamara
- Email: [email protected]
- Phone: +44 20 7040 8154
- City, University of London
- Northampton Square
- London EC1V 0HB. United Kingdom