TL;DR
This study experimentally investigates the impact of landslide-generated impulse waves on gravity dams using a 3D physical model. It reveals that landslide volume and angle are critical for wave amplitude, and establishes relationships between wave characteristics and dam surface pressure, providing crucial data for dam safety assessments.
What Problem Does This Paper Address?
Mountainous regions often feature reservoirs, and the construction and operation of these reservoirs can destabilize slopes, leading to landslides. When a landslide occurs near a dam, it can generate powerful impulse waves that pose a significant threat to the dam's structural integrity and the safety of downstream communities. Existing methods for predicting the pressure these waves exert on dams are insufficient, especially for complex reservoir environments. This paper aims to fill this gap by experimentally characterizing the pressure distribution and variation on a gravity dam's surface under the impact of landslide-generated impulse waves, offering a more accurate basis for dam safety evaluations and risk assessments.
Methods
The researchers conducted a series of three-dimensional physical model tests using a large flume. The experimental setup included a gravity dam model and a landslide simulation device. To mimic real-world rock landslides, they used five different-sized blocks to simulate the landslide material. The tests involved varying landslide angles (30°, 40°, 50°, 60°) and landslide volumes (0.4 m³, 0.6 m³, 0.9 m³). They employed ultrasonic wave-measuring instruments to record water level changes and a wave pressure acquisition system to measure dynamic wave pressure on the dam surface. Pressure sensors were embedded at various heights on the dam's center section. The study focused on analyzing the wave height propagation and the pressure characteristics at the dam's center.
Datasets
The study utilized data generated from their physical model experiments. They conducted 16 different test conditions by varying landslide parameters (volume, angle, dimensions). The data collected included instantaneous water levels and dynamic wave pressures on the dam surface at different depths. No external datasets were used for training or evaluation.
Key Insights and Ideas
- Landslide Parameters Matter: The landslide volume and angle are identified as the primary drivers controlling the amplitude of the initial impulse wave impacting the dam.
- Pressure Distribution is Non-Uniform: The wave pressure on the dam surface is not uniform with depth. It's generally smaller in the middle and larger at the top and bottom, with a minimum pressure occurring at approximately two-thirds of the water depth.
- Wave Amplitude Dictates Pressure: The maximum wave pressure on the dam surface exhibits a power-law relationship with the maximum amplitude of the impulse wave.
- Total Pressure is Linked to Amplitude: The total pressure exerted by the impulse wave on the dam surface is strongly influenced by the initial wave amplitude, showing a quadratic relationship.
- 3D Effects are Significant: The study suggests that the surge degree in a 3D model is larger than in a 2D model, highlighting the importance of considering three-dimensional effects in such analyses.
Key Results
- The dimensionless first wave amplitude (Am/ho) ranged from 0.017 to 0.043.
- The maximum wave pressure generally occurred near the water surface, with a minimum around two-thirds of the water depth (z/zo = 0.673).
- A power function was derived to express the dimensionless pressure near the water surface (P/yho) as a function of the dimensionless first wave amplitude (Am/ho): P/yho = 0.34(Am/ho)⁰.⁶¹⁵ (R² = 0.979).
- A formula was developed to predict the total pressure per unit width of the impulse waves on the dam surface (Fm/(yho²)) based on the first wave amplitude: Fm/(yho²) = -3.72(Am/ho)² + 0.95(Am/ho) (R² = 0.986).
- The study found that previous formulas for granular landslides might not accurately predict impulse wave amplitudes for block-composed landslides, necessitating a re-fitting of the equation.
Limitations and Future Work
The authors acknowledge that the experimental results for the initial wave amplitude showed significant deviations from previous calculations based on granular landslides, suggesting that their findings are specific to the block-composed landslide simulation used. They also note that the interference signal at the bottom of the dam was not entirely eliminated. Future work could involve exploring a wider range of landslide materials and geometries, investigating the long-term effects of repeated wave impacts, and further refining the predictive models for dam safety assessments.
Summary updated