在多物理场AI助手MOOSEnger中部署前沿智能体技术

中文摘要 AI 辅助

多物理场面向对象仿真环境（MOOSE）是用于构建多物理场仿真应用的开源有限元框架。有效使用多物理场环境需要专业知识，这对许多领域科学家和工程师构成了障碍。爱达荷国家实验室（INL）开发的MOOSEnger是一款面向MOOSE框架的、具备工具调用能力的领域专用AI智能体。本研究为MOOSEnger扩展了一个专注于本地托管模型的管控层，该管控层为智能体提供了完整流程：从MOOSE知识库中检索上下文知识，通过与仿真可执行环境交互验证并诊断生成的输入，以及从交互中提取经验并存储到持久化内存中。该框架在国家反应堆创新中心虚拟测试台（VTB）的一个工程问题上得到验证，展现出支持实际多物理场仿真工作流的潜力。此外，针对扩散、纳维-斯托克斯、相场、塑性、多孔介质流动、固体力学、瞬态传热及反应堆网格生成等不同类别，每类包含25个提示/案例，对智能体性能进行了评估。我们将MOOSEnger-Gemma4与MOOSEnger-GPT-5.2进行对比，同时与不具备智能体能力的基准模型Gemma4和GPT-5.2对比。MOOSEnger-GPT-5.2表现出微弱优势，成功率达90%，而MOOSEnger-Gemma4的成功率为76.5%；基准模型的表现差得多，GPT-5.2仅为5%，Gemma4为0%，凸显了智能体管控层的重要作用。

英文摘要

The Multiphysics Object-Oriented Simulation Environment (MOOSE) is an open-source finite-element framework for building multiphysics simulation applications. Using a multiphysics environment effectively demands specialized expertise, creating a barrier for many domain scientists and engineers. MOOSEnger, developed at Idaho National Laboratory (INL), is a domain-specific, tool-enabled AI agent built for the MOOSE Framework. This work extends MOOSEnger with a harness focused on locally-hosted models. The harness gives the agent a full pipeline: it retrieves contextual knowledge from the MOOSE repository, validates and diagnoses the resulting input through interaction with the simulation executable environment, and extracts and stores lessons in a persistent memory. The resulting framework is demonstrated on an engineering problem from the National Reactor Innovation Center Virtual Test Bed (VTB), illustrating its potential to support realistic multiphysics simulation workflows. Additionally, the agent performance is evaluated on different categories including diffusion, Navier--Stokes, phase field, plasticity, porous media flow, solid mechanics, transient heat transfer, and reactor mesh generation. Each category consists of 25 prompts/cases. We compare MOOSEnger-Gemma4 against MOOSEnger-GPT-5.2, alongside baseline Gemma4 and GPT-5.2 without agentic capabilities. MOOSEnger-GPT-5.2 shows a slight edge, achieving a 90\% success rate versus 76.5\% for MOOSEnger-Gemma4. The baseline models perform far worse, at just 5\% (GPT-5.2) and 0\% (Gemma4), underscoring the impact of the agentic harness.

发表机构

• Texas A&M University (TAMU)（德克萨斯农工大学）
• Idaho National Laboratory (INL)（爱达荷国家实验室）

机构由 AI 辅助整理，请以论文原文为准。