用大模型自动完成流体模拟全流程，成功率超80%
ChatCFD: An LLM-Driven Agent for End-to-End CFD Automation with Structured Knowledge and Reasoning
- 构建多智能体系统，融合领域知识与错误定位实现自主求解
- 在315个测试案例中成功率达82.1%，物理合理性达68.12%
- 适合科研人员快速复现复杂流体问题，降低仿真门槛
计算流体力学（CFD）对科学进步至关重要，但受操作复杂性和高专业门槛制约。本文提出ChatCFD，一个基于大语言模型（LLM）的多智能体系统，用于端到端自动化OpenFOAM仿真。依托DeepSeek-R1/V3，ChatCFD整合结构化领域知识库、精准错误定位器和迭代反思机制，在315个基准案例中实现82.1%的执行成功率（远超MetaOpenFOAM的6.2%和Foam-Agent的42.3%），物理保真度达68.12%（新指标，衡量科学合理性）。专用物理解释器实现97.4%的摘要保真度，平衡表述流畅性与物理约束。资源分析显示，平均消耗192.1k tokens与$0.208/案例，显著低于基线。消融实验表明，错误定位器与求解器模板数据库为关键组件，后者移除后准确率降至48%。系统在自主求解器选择上达95.23%成功率，湍流建模100%成功，并成功复现文献中的复杂案例（如NACA0012、超音速喷管），在基线失败区域实现60-80%成功率。模块化且兼容MCP设计，支持可扩展的协作式AI-CFD。代码已开源：https://github.com/ConMoo/ChatCFD
原文摘要 · Abstract (English)
Computational Fluid Dynamics (CFD) is critical for scientific advancement but is hindered by operational complexity and high expertise barriers. This paper introduces ChatCFD, a Large Language Model (LLM)-driven multi-agent system designed for end-to-end CFD automation using OpenFOAM. Powered by DeepSeek-R1/V3, ChatCFD integrates structured domain knowledge bases, a precise error locator, and iterative reflection to dramatically outperform existing methods. On 315 benchmark cases, ChatCFD achieves 82.1% execution success (vs. 6.2% for MetaOpenFOAM and 42.3% for Foam-Agent) and 68.12% physical fidelity - a novel metric assessing scientific meaningfulness beyond mere runnability. A dedicated Physics Interpreter attains 97.4% summary fidelity, bridging the gap between narrative fluency and the enforcement of tight physical constraints. Resource analysis confirms efficiency, averaging 192.1k tokens and $0.208 per case, significantly lower than baseline costs. Ablation studies identify the Error Locator and Solver Template DB as critical, with the latter's removal collapsing accuracy to 48%. The system exhibits robust flexibility, achieving 95.23% success in autonomous solver selection and 100% in turbulence modeling, while successfully reproducing complex literature cases (e.g., NACA0012, supersonic nozzle) with 60-80% success rates where baselines failed. Featuring a modular, MCP-compatible design, ChatCFD facilitates scalable, collaborative AI-driven CFD. Code is available at: https://github.com/ConMoo/ChatCFD
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。