Open Access

ARTICLE

A Fully Lagrangian Mesh-Free Framework for Fluid–Structure Interaction Based on WC-MPS and Hybrid TL–UL Formulations

Saeed Tavakoli^*, Ahmad Shakibaeinia, Najib Bouaanani

Department of Civil, Geological, and Mining Engineering, Polytechnique Montréal, Montréal, QC, Canada

* Corresponding Author: Saeed Tavakoli. Email:

(This article belongs to the Special Issue: Recent Developments in SPH and CFD Methods for Complex Flow Simulations)

Computer Modeling in Engineering & Sciences 2026, 147(3), 16 https://doi.org/10.32604/cmes.2026.081925

Received 11 March 2026; Accepted 15 May 2026; Issue published 30 June 2026

Abstract

Fluid–structure interaction (FSI) plays a critical role in civil engineering applications, directly influencing structural safety, resilience, and performance. However, the inherent multiphysics complexity of FSI problems presents significant challenges for numerical modeling, particularly under highly dynamic flow conditions. This study presents a fully Lagrangian mesh-free framework for FSI based on the moving particle semi-implicit (MPS) method. The approach couples an enhanced weakly compressible MPS (WC-MPS) fluid solver with a hybrid total–updated Lagrangian (TL–UL) MPS formulation for elastic solids. In the solid phase, strains are evaluated in the reference configuration, while momentum balance is enforced in the current configuration, ensuring consistency under large deformations. The framework incorporates corrected kernel gradients and rotationally consistent particle interactions to suppress spurious zero-energy modes. The fluid solver includes artificial density diffusion and particle regularization to enhance stability and pressure smoothness. Stable and conservative coupling between phases is achieved through a normal-flux-based interface treatment combined with a velocity-consistent particle-grouping strategy. Furthermore, an improved interface particle registration criterion is introduced to ensure accurate and efficient boundary detection at deforming interfaces. The current implementation and validation of the proposed framework are restricted to two-dimensional (2D) FSI scenarios. The framework is validated against four benchmark problems: the dynamic response of a cantilever beam, dam-break flow, the Turek-Hron FSI benchmark (flow around a flexible flag), and a dam-break flow interacting with an elastic gate. The results show strong agreement with analytical, experimental, and numerical references, demonstrating the robustness and stability of the formulation for hydro-elastic interactions. Ultimately, the framework provides a simple, unified, and reliable tool for simulating large-deformation FSI problems in a fully Lagrangian context.

---

Keywords

---