Slope with Predetermined Shear Plane Stability Predictions under Cyclic Loading with Innovative Time Series Analysis...

Area of Science:

• Geotechnical Engineering
• Computational Mechanics
• Seismology

Background:

• Slope stability analysis is crucial for infrastructure safety, especially under seismic conditions.
• Cyclic seismic loads can induce strain softening and lead to landslides.
• Predicting slope stability coefficients requires understanding complex soil behavior under dynamic loading.

Purpose of the Study:

• To develop a mechanical learning method for predicting slope stability coefficients under cyclic seismic loads.
• To investigate the influence of cyclic loading parameters on soil strain softening and fracture propagation.
• To establish a predictive model for slope stability coefficients based on time-series data.

Main Methods:

• Simulating cyclic shear tests on soil specimens to observe strain softening.
• Applying Sarma's limit equilibrium method to analyze stability coefficients under varying cyclic loads.
• Utilizing Box-Jenkins modeling for time-series prediction of stability coefficients via mechanical learning.

Main Results:

• Coordination number reduction accurately indicates strain softening.
• Shear stress reduction and cyclic stress amplitudes influence fracture zone penetration and strain softening.
• Confining pressure has a limited effect on inhibiting strain softening.
• Mechanical learning accurately predicts stability coefficient patterns without loading.

Conclusions:

• The proposed mechanical learning method offers a reliable approach for predicting slope stability coefficients.
• Understanding strain softening indicators and the impact of cyclic loading is vital for landslide prediction.
• This research bridges geological landslide investigation with theoretical stability prediction.