This work presents a novel computational modeling framework that tackles the challenges of fluid–structure interaction (FSI) problems. FSI has a wide range of applications—from analyzing the aeroelasticity of aircraft wings to studying blood circulation in aortic valves. However, solving FSI problems is challenging because of the complex, time-varying geometry at the fluid–structure interface. It is particularly problematic for grid-based methods to track the interface and update meshes when faced with large deformations and structural failures. To address these issues, we propose a meshfree framework that combines smoothed particle hydrodynamics (SPH) and peridynamics (PD). This approach not only simplifies the discretization process but also offers superior effectiveness and robustness in dealing with FSI problems involving large deformations and failures. Our proposed framework has the potential to enhance the simulation of FSI problems in a variety of fields, ultimately leading to improved designs and outcomes.

The proposed meshfree framework employs weakly compressible SPH to model the fluid phase, while PD are used to model the solid materials. The SPH–PD framework involves a partitioned coupling procedure, in which the SPH moving ghost particles serve as a medium for data transfer. Violent free-surface flow interacting with structures are captured with the proposed method. To yield accurate simulations of flows in irregular channels, we develop a new periodic boundary condition (PBC) algorithm. In our algorithm, we define an inlet PBC zone and transfer some particles in the PBC zone upstream to fill the empty zone created by downstream particle motion. Our algorithm is demonstrated to be robust for FSI modeling in both regular and irregular channels. We also develop a novel single-phase surface tension model for SPH, which can simulate droplet dynamics based on a fast detection algorithm of free surface. Our method provides an effective solution to those FSI problems involving surface tension, such as droplets interacting with deformable structures.

This work also explores the potential of the SPH–PD framework in biomedical engineering scenarios, specifically in modeling the deformation and rupture of blood vessels, along with the use of both Fung-type hyperelasticity and Casson’s non-Newtonian fluid models. The proposed method successfully captures an essential physical phenomenon of blood pressure–induced spontaneous ruptures of blood vessels. Moreover, this study systematically characterizes the effects of various factors, such as material constitutive models, loads from surrounding tissues, the off-axis distance of an aneurysm, and outlet resistance, on the deformation and damage behaviors of blood vessels. The results indicate that surrounding tissues significantly influence blood vessel deformation and damage behaviors, underscoring the importance of considering these aspects in future biomechanic simulations. Further, this study suggests that the SPH–PD framework can be extended to 3D cases and applied to other engineering applications, such as soft swimming robots and patient-specific biomedical modeling.

To enhance the computational efficiency of meshfree methods, particularly SPH, we study adaptive particle refinement (APR) strategies. Our validation results demonstrate that the proposed APR algorithm not only improves simulation accuracy but also offers flexibility in implementation, making it a valuable tool for researchers across a range of disciplines.