TL;DR
This paper introduces an improved numerical model for simulating backward erosion piping (BEP) in dikes. By incorporating unsaturated seepage within the dike body and coupling it with seepage-stress effects, the model more accurately predicts piping evolution and its significant impact on dike slope stability, showing a substantial reduction in safety factors.
What Problem Does This Paper Address?
Backward erosion piping (BEP) is a critical failure mechanism for water-retaining structures like dikes, often leading to catastrophic collapses. While existing models can simulate piping, they often overlook the crucial influence of seepage within the dike body itself, focusing primarily on foundation seepage. This oversight limits their accuracy in predicting the progression of piping channels and their subsequent impact on dike stability. This paper aims to bridge this gap by developing a more comprehensive model that accounts for both dike body seepage and the coupled effects of seepage and soil stress.
Methods
The core of this research is an improved numerical model for backward erosion piping (BEP), building upon Wewer's existing model. The key innovations are:
- Incorporation of Unsaturated Seepage in the Dike Body: The model now solves for the seepage field within the dike body using Darcy's law and the continuity equation for unsaturated porous media. This is a significant departure from previous models that primarily focused on foundation seepage.
- Seepage-Stress Coupling: The model accounts for the interaction between pore water pressure and the soil's stress state. This is achieved by using effective stress in the equilibrium equation and incorporating constitutive models (Drucker-Prager criterion) to capture soil deformation and potential plastic strain.
- Validation: The improved model's accuracy was validated against experimental data from the IJkdijk experiment and compared with Wewer's original model results.
The numerical simulations were performed using COMSOL Multiphysics with a finite element method approach.
Datasets
The primary dataset used for validation was from the IJkdijk experiment. This experiment involved a scaled model of a dike and foundation, providing data on piping development and pore pressure changes. The paper also references Wewer's previous work and experimental data for comparison.
Key Insights and Ideas
- Dike Body Seepage Matters: The study strongly suggests that neglecting seepage within the dike body leads to less accurate predictions of piping evolution. The improved model, by including this factor, shows a more consistent development of piping length and pore pressure changes compared to experimental data.
- Seepage-Stress Coupling is Crucial for Stability Analysis: The interplay between pore water pressure and soil stress is vital for understanding how piping affects the overall stability of the dike slope. This coupling allows for the simulation of deformation and the identification of critical failure zones.
- Piping Dramatically Reduces Dike Stability: The formation and progression of a piping channel significantly compromise the dike's structural integrity, leading to a substantial decrease in its safety factor. This highlights BEP as a direct threat to dike performance.
- Water Level Dynamics Influence Piping: The study also explored the impact of water level changes, demonstrating that while a drop in water level can slow down or halt piping development due to reduced hydraulic gradients, it can also lead to upstream slope instability due to differential water pressures.
Key Results
- The improved piping model demonstrated higher accuracy in simulating pipe length development compared to Wewer's model, aligning more closely with the IJkdijk experimental results.
- The model successfully predicted a significant decrease in pore pressure at monitoring points as the piping channel developed, a phenomenon consistent with experimental observations.
- The analysis showed that the formation of a complete erosion channel led to a sharp drop in the dike slope's safety factor, from approximately 1.54 to 1.488, indicating a critical loss of stability.
- Under simulated water level drops, the model showed that piping could cease development due to insufficient hydraulic gradients, but this could also induce instability on the upstream slope.
Limitations and Future Work
The authors acknowledge several limitations:
- The improved model does not consider the collapse of soil above the pipe during BEP.
- The study did not comprehensively assess the impact of BEP on dike stability by considering changes in soil parameters such as Young's modulus.
Future research could focus on incorporating these aspects to further enhance the model's predictive capabilities. Additionally, further investigation into the complex interplay of soil parameters and BEP, as well as more detailed analysis of dike stability under various BEP scenarios, is suggested.
Summary updated