Agentic AI for Particle-Based Simulation: Automating SPH Workflows for Debris Flow Modeling

2026-05-09

AI Summary

This work addresses the challenges of deploying mesh-free particle methods—such as Smoothed Particle Hydrodynamics (SPH)—in engineering simulations, which are often hindered by complex setup procedures, parameter sensitivity, and difficulties in result interpretation. The paper presents the first large language model (LLM) agent framework tailored for end-to-end SPH simulation, integrating multimodal inputs (text and sketches), orchestrated computational tools, and a human-in-the-loop mechanism to automate the entire workflow from modeling to post-processing. Innovatively, it introduces a cognitive task evaluation framework coupled with domain-aware physical reasoning, substantially lowering user operational barriers and failure rates. Demonstrated on debris flow simulations, the approach exhibits exceptional robustness in configuration, high-quality visualization, and effective data extraction capabilities.

Abstract

Physics-based simulation underpins engineering analysis but remains difficult to deploy in practice due to complex setup, parameterization, and interpretation. While Large Language Model-based agentic systems have shown promise in automating engineering computing workflows, they have primarily targeted structured, mesh-based problems. We present the first agentic AI workflow for meshless simulation in computational mechanics, demonstrated on debris flow modeling using Smoothed Particle Hydrodynamics (SPH) with the software DualSPHysics. By integrating tool orchestration, multimodal inputs (text and sketches), and human-in-the-loop interaction, the framework enables end-to-end simulation workflows for a class of problems that are inherently less structured and more challenging to automate. Results show that multimodal inputs not only enhance user experience but also reduces failure modes over text-only descriptions. Human-in-the-loop is critical for resolving ambiguities and handling SPH-specific configurations. We further introduce a cognitive-task-based evaluation of post-processing, showing strong performance in visualization and data extraction, with remaining gaps in higher-level SPH-specific physical reasoning that are amenable to improvement through domain-aware modeling. These results establish the viability of agentic AI for particle-based simulation and underscore its potential to transform the accessibility and efficiency of computational mechanics workflows.

Problem

• particle-based simulation
• debris flow modeling
• Smoothed Particle Hydrodynamics
• computational mechanics
• workflow automation

Innovation

• Agentic AI
• Smoothed Particle Hydrodynamics (SPH)
• meshless simulation
• multimodal input
• human-in-the-loop