Towards an AI Fluid Scientist: LLM-Powered Scientific Discovery in Experimental Fluid Mechanics

2025-12-04

AI Summary

Current experimental fluid dynamics lacks AI systems capable of end-to-end autonomous scientific discovery. Method: This work introduces the “AI Fluid Scientist” framework—a large language model–driven multi-agent system integrating virtual–physical interaction, human-in-the-loop collaboration, programmable recirculating water tunnel control, robotic experimental platforms, and neural-network-based physical modeling. It enables a closed-loop workflow spanning hypothesis generation, experimental design, automated execution, data analysis, and scientific paper drafting. Contributions/Results: To our knowledge, this is the first demonstration of fully autonomous, end-to-end scientific discovery in experimental fluid mechanics. Neural-network–based physical law fitting achieves 31% higher accuracy than polynomial regression. The system successfully reproduces canonical results (e.g., lock-in error <4%), accurately captures critical spacing trends, and conducts hundreds of autonomous experiments—substantially enhancing research efficiency, reproducibility, and scalability.

Technology Category

• Humans and AI: Game Design — Virtual Humans, NPCs and Autonomous Characters
• Multiagent Systems: Teamwork
• Cognitive Modeling & Cognitive Systems: Agent Architectures

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Agentic search

Abstract

The integration of artificial intelligence into experimental fluid mechanics promises to accelerate discovery, yet most AI applications remain narrowly focused on numerical studies. This work proposes an AI Fluid Scientist framework that autonomously executes the complete experimental workflow: hypothesis generation, experimental design, robotic execution, data analysis, and manuscript preparation. We validate this through investigation of vortex-induced vibration (VIV) and wake-induced vibration (WIV) in tandem cylinders. Our work has four key contributions: (1) A computer-controlled circulating water tunnel (CWT) with programmatic control of flow velocity, cylinder position, and forcing parameters (vibration frequency and amplitude) with data acquisition (displacement, force, and torque). (2) Automated experiments reproduce literature benchmarks (Khalak and Williamson [1999] and Assi et al. [2013, 2010]) with frequency lock-in within 4% and matching critical spacing trends. (3) The framework with Human-in-the-Loop (HIL) discovers more WIV amplitude response phenomena, and uses a neural network to fit physical laws from data, which is 31% higher than that of polynomial fitting. (4) The framework with multi-agent with virtual-real interaction system executes hundreds of experiments end-to-end, which automatically completes the entire process of scientific research from hypothesis generation, experimental design, experimental execution, data analysis, and manuscript preparation. It greatly liberates human researchers and improves study efficiency, providing new paradigm for the development and research of experimental fluid mechanics.

Problem

• Developing an autonomous AI framework for experimental fluid mechanics
• Validating the framework using vortex-induced vibration studies
• Automating the entire scientific workflow from hypothesis to manuscript

Innovation

• AI framework autonomously executes complete experimental workflow
• Human-in-the-Loop discovers phenomena and fits laws with neural network
• Multi-agent system runs hundreds of experiments end-to-end automatically