A coupled DDA-MPM method for the collapse-water inrush chain disaster modeling in underwater tunnels

Tunnels constructed beneath reservoirs face a high risk of water inrush and surrounding rock collapse during construction. Clarifying the disaster mechanism is crucial for safe tunnel construction. However, traditional single numerical methods often struggle to accurately simulate the complex fluid-rock interaction processes involved in such disasters, and the simulation results fail to provide effective guidance. This paper develops a fluid–structure interaction numerical method based on the coupling of the discontinuous deformation analysis (DDA) and the material point method (MPM) to systematically investigate the chain mechanism of surrounding rock collapse and water inrush disasters in underwater tunnels. Firstly, an MPM for simulating weakly compressible Newtonian fluids was established, and the contact forces between the fluid and blocks were calculated using the penalty spring method to achieve bidirectional coupling between DDA and MPM. Subsequently, the coupled method was systematically validated through typical tests such as block collapse, dam break, and wave generation. The proposed DDA-MPM method captures not only block motion and fluid evolution but also their mutual interactions, achieving a level of accuracy that single numerical methods cannot reach. Finally, taking the water inrush accident in the Shijingshan Tunnel as the engineering background, the coupled method was applied to analyze the tunnel construction-induced surrounding rock collapse and water inrush. This revealed the spatiotemporal evolution mechanisms of rock mass instability and collapse, the formation of water inrush channels, and the migration of water within the tunnel network. The research indicates that tunnel excavation unloading leads to the overall instability and collapse of the rock mass in the weathered zone, forming a connecting channel between the reservoir and the tunnel, which subsequently triggers water inrush and its diffusion to adjacent tunnels. This study confirms that the proposed method effectively simulates fluid–structure interaction disaster problems, thereby laying a theoretical groundwork for the early detection, prevention, and control of analogous engineering disasters.