HGATSolver: A Heterogeneous Graph Attention Solver for Fluid-Structure Interaction

2026-01-14

AI Summary

Existing learning-based solvers struggle to uniformly model the heterogeneous dynamics of fluid and solid regions in fluid–structure interaction (FSI) systems, often yielding unstable predictions due to the complex coupling at interfaces and disparate learning difficulties across domains. To address this, this work proposes a heterogeneous graph neural network framework that explicitly distinguishes fluid, solid, and interface node types, and employs domain-specific message-passing mechanisms. The approach further incorporates physics-informed gating—implemented as a learnable adaptive relaxation factor—and a cross-domain gradient-balancing loss function weighted by predictive uncertainty. Evaluated on two newly constructed FSI benchmarks and one public dataset, the method achieves state-of-the-art performance, offering an effective paradigm for high-fidelity surrogate modeling of multiphysics-coupled systems.

Technology Category

• Machine Learning: Neuro-Symbolic Learning
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Multiagent Systems: Adversarial Agents

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Web applications in cross-disciplinary domains and verticals such as mixed reality, smart cities, and digital health
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Fluid-structure interaction (FSI) systems involve distinct physical domains, fluid and solid, governed by different partial differential equations and coupled at a dynamic interface. While learning-based solvers offer a promising alternative to costly numerical simulations, existing methods struggle to capture the heterogeneous dynamics of FSI within a unified framework. This challenge is further exacerbated by inconsistencies in response across domains due to interface coupling and by disparities in learning difficulty across fluid and solid regions, leading to instability during prediction. To address these challenges, we propose the Heterogeneous Graph Attention Solver (HGATSolver). HGATSolver encodes the system as a heterogeneous graph, embedding physical structure directly into the model via distinct node and edge types for fluid, solid, and interface regions. This enables specialized message-passing mechanisms tailored to each physical domain. To stabilize explicit time stepping, we introduce a novel physics-conditioned gating mechanism that serves as a learnable, adaptive relaxation factor. Furthermore, an Inter-domain Gradient-Balancing Loss dynamically balances the optimization objectives across domains based on predictive uncertainty. Extensive experiments on two constructed FSI benchmarks and a public dataset demonstrate that HGATSolver achieves state-of-the-art performance, establishing an effective framework for surrogate modeling of coupled multi-physics systems.

Problem

• fluid-structure interaction
• heterogeneous dynamics
• domain coupling
• learning-based solver
• multi-physics systems

Innovation

• Heterogeneous Graph Neural Network
• Fluid-Structure Interaction
• Physics-Informed Learning
• Adaptive Gating Mechanism
• Gradient Balancing