The tsunamis induced by rigid landslides such as ice and rock may affect the safety of nearshore floating structures, since the generated waves tend to be strongly non-linear. In this work, a fully Lagrangian meshless method MPS-DEM is developed to solve the fluid-solid interaction problem. The improved moving particle semi-implicit (MPS) method is applied to simulate the incompressible fluid flow, while the Discrete Element Method (DEM) is used to build the solid-ramp interaction. The MPS-DEM coupled method is applied to simulate the subaerial rigid landslide-tsunamis. Numerical results agree well with experimental results.
Landslides often occurs on the ramps near the ocean, rivers and reservoirs. Blocking of the rivers by the falling stones leads to dramatic change of environment. Besides, non-linear waves induced by landslides may pose a threaten to the safety of nearshore structures and people. Therefore, it is necessary to investigate the movement of landslides and the propagation of waves.
Model experiment is one of mainly approach to evaluate the influence caused by landslides. Heinrich (1992) conducted experiments to study the surge induced by submarine and aerial landslides. The wave profile and the trajectory of solid bodies was investigated. Lin et al. (2015) experimentally investigated the interaction between the landslide-induced surge wave and a dam. Heller et al. (2016) carried out a 3-D experiment to investigate the surge induced by subaerial landslides. These model experiments mentioned above are usually used to verify the accuracy of numerical methods.
With the development of the computer hardware, many numerical techniques have been proposed for the simulation of complex flows. Particle-based methods show their superiority to capture the free-surface with large deformation. Weakly Compressible Smoothed Particle Hydrodynamics (WCSPH) method was firstly proposed by Gingold and Monaghan (1977) for the problem of astrophysics. Then, SPH was widely applied for the problem of violent flows. In a few decades, Koshizuka and Oka (1996) developed Moving Particle Semi-implicit (MPS) method to simulated the incompressible flows. There were some draw backs for particle-based methods, such as pressure oscillation and low computational efficiency. High order schemes (Khayyer et al.,2011; Liu et al., 2019; Duan et al., 2021) for gradient model and Laplacian model were developed to enhance the performance of particle-based method. Particle Shifting Technique (PST) (Xu et al, 2009; Khayyer et al, 2017; Duan et al., 2018; Sun et al, 2019a) was also proposed to avoid errors and numerical oscillation bringed by uneven distribution of particles. With the help of multi-resolution techniques (Sun et al., 2019b), parallel technique and GPU acceleration technique (Xie et al., 2020), the computational efficiency was improved obviously. With the efforts of several researchers, particle-based methods have been applied to solve more complicated problems, such as multi-phase flows (Khayyer et al., 2019, Wen et al., 2021, Xie et al., 2021b), fluid-structure interaction (Khayyer et al., 2021; Gotoh et al, 2021; Zhang et al., 2021b; Xie et al., 2021a) and flows passing through porous structures (Wen et al., 2018; Khayyer et al., 2018).
