Agentic AI 用于分阶段三维有限元隧道建模

中文摘要 AI 辅助

三维有限元（FE）分析能最全面地呈现隧道开挖引起的地表移动，但过程缓慢：工程师必须操作 PLAXIS 3D 界面，完成项目设置、地层划分、材料定义、结构构件、网格划分和分阶段计算。本文提出一种流程，其中大语言模型（LLM）通过模型上下文协议（MCP）驱动 PLAXIS 3D，无需人工操作界面。一个专门构建的 MCP 服务器将官方远程脚本 API 暴露为 19 个类型化工具。四个智能体分担工作：编排者、几何、计算和验证。它们利用 15 个可检查的领域知识技能模块和一个带来源标记的输入阶段。验证智能体设有信息隔离，在网格划分前检查点以及作为计算门控时，依据原始规范审计每个模型。该系统在曼谷地下土层的 12 个单隧道问题上，跨越四个能力层级进行了评估，使用 15 臂消融矩阵；执行了 13 个自动化臂和一个手动专家基线，每个单元运行一次。完整系统构建了 12 个模型中的全部 12 个，平均耗时 6.1 分钟，且每个模型均通过干净审计；相同模型手动构建需 24.6 分钟。领域技能至关重要：在完整架构但所有 15 个模块被 withheld 的情况下，未生成任何模型。基础模型能力表现为阈值而非梯度。Fable 5、Opus 5 和 Sonnet 5 各自构建了 12 个中的 12 个，而 Haiku 4.5 在所有 12 个上均停止。在成功构建的模型中，一致性精确：111 个可比单元中的 107 个复现了参考网格至单元级别，86 个已求解单元中的 83 个与手动构建的专家模型在最大沉降上差异在 1.01% 以内，在槽宽参数上差异在 1.21% 以内。每个异常都可追溯到输入阶段的值误读。因此，智能体有限元自动化中的可靠性问题在于规范读取，而非代码生成。

英文摘要

Three-dimensional finite-element (FE) analysis gives the most complete picture of tunnelling-induced ground movement, but it is slow: an engineer must drive the PLAXIS 3D interface through project set-up, stratigraphy, materials, structural elements, meshing and staged calculation. This paper presents a pipeline in which a large language model (LLM) drives PLAXIS 3D through the Model Context Protocol (MCP), with no human operating the interface. A purpose-built MCP server exposes the official remote-scripting API as 19 typed tools. Four agents share the work: orchestrator, geometry, calculation and verification. They draw on 15 inspectable domain-knowledge skill modules and a provenance-tagged intake stage. The verification agent is information-barriered, auditing every model against the raw specification at a pre-mesh checkpoint and again as a calculation gate. It was evaluated on 12 single-tunnel problems in Bangkok subsoil across four capability tiers, using a 15-arm ablation matrix; 13 automated arms and a manual expert baseline were executed, one run per cell. The full system built 12 of 12 models at a mean of 6.1 min with a clean audit on every one; the same models took 24.6 min by hand. Domain skills were decisive: with the complete architecture but all 15 modules withheld, no model was produced. Base-model capability acted as a threshold rather than a gradient. Fable 5, Opus 5 and Sonnet 5 each built 12 of 12, while Haiku 4.5 halted on all 12. Among models that built, agreement was exact: 107 of 111 comparable cells reproduced the reference mesh to the element, and 83 of 86 solved cells agreed with the hand-built expert model within 1.01 % in maximum settlement and 1.21 % in trough-width parameter. Every exception traced to a value misread at intake. The reliability problem in agentic FE automation therefore lies in specification reading, not in code generation.

发表机构

• Asian Institute of Technology（亚洲理工学院）
• Geotechnical and Foundation Engineering Company Limited（岩土与基础工程有限公司）
• AI Research Group, Department of Civil Engineering, King Mongkut’s University of Technology Thonburi (KMUTT)（孔敬理工大学（KMUTT）土木工程系人工智能研究组）
• GOE Consultants Company Limited（GOE咨询有限公司）

机构由 AI 辅助整理，请以论文原文为准。