Probabilistic Seismic Stability Analysis of Reinforced Soil Retaining Structures in Partially Saturated Conditions

• Published in: International journal of geomechanics 2024-08, Vol.24 (8)
• Main Authors: Agarwal, Ekansh, Pain, Anindya, Annapareddy, V. S. Ramakrishna, Zhou, Annan
• Format: Article
• Language: English
• Subjects: Backfill; Constraints; Design; Partially saturated soils; Probabilistic methods; Probability theory; Quadratic programming; Randomness; Reinforced soils; Response surface methodology; Seismic stability; Seismicity; Soil; Soil structure; Soils; Stability analysis; Stochasticity; Structures; Technical Papers
• ISSN: 1532-3641; 1943-5622
• DOI: 10.1061/IJGNAI.GMENG-9445

Abstract The conventional design of reinforced soil retaining structures (RSRS) assumes a dry or fully saturated backfill in a deterministic approach, which may not reflect real-field conditions accurately. While there are ongoing efforts to integrate partially saturated conditions, there is a noticeable lack of emphasis on simplifying the design processes associated with these conditions. This constraint restricts their practical application, particularly in the context of probabilistic scenarios. This study employed the collocation-based stochastic response surface method to present a simple probabilistic method for the seismic stability analysis of reinforced soil retaining structures in partially saturated conditions. To improve accuracy, the horizontal slice is modified to include the effects of cohesion and suction stress in no-flow and infiltration conditions with the least number of assumptions. For practical applicability, the whole process is made efficient by using the nonlinear constrained sequential quadratic programming. To elaborate, a probabilistic parametric study is conducted on a typical reinforced soil structure with silty sand backfill, a recommended material for its construction. The randomness associated with each system, that is, soil, hydraulic parameters, and seismicity, is considered. The results emphasize the need to consider the probabilistic aspect while designing the reinforced soil structures, especially when dealing with partially saturated conditions. The research promotes the adoption of simplified probabilistic methods to encourage field engineers to re-evaluate designs that were originally calculated using deterministic methods. Practical Applications When designing the soil retaining structures, engineers usually assume the soil is either completely dry or completely wet. However, this may not match real-world conditions. While there are efforts to account for partially saturated conditions, the methods are often complicated. This makes it hard to use them in real-life situations, especially when considering the chance of earthquakes. This becomes more complicated when it comes to probabilistic design that involves considering uncertainties and variations in soil and environmental conditions. Even though engineers have been trying out new ways of predicting things in engineering, the industry is still a bit unsure about using these methods fully in their designs. This may render the structures prone to failure in case