📄 中文摘要
研究提出了一种结合大型语言模型（LLMs）、检索增强生成（RAG）和基于图的RAG（GraphRAG）的方法，用于生成设计结构矩阵（DSMs）。在两个不同的应用案例中进行测试，分别是电动螺丝刀和具有已知架构参考的CubeSat，评估其在确定预定义组件之间关系和识别组件及其后续关系这两项关键任务中的表现。通过评估DSMs的每个元素及整体架构，测量其性能。尽管面临设计和计算挑战，研究发现了自动化DSMs生成的机会，所有代码均公开以便于重现和进一步研究。
基于检索增强（知识图谱）和大型语言模型驱动的网络物理系统设计结构矩阵生成
📄 English Summary
Retrieval Augmented (Knowledge Graph), and Large Language Model-Driven Design Structure Matrix (DSM) Generation of Cyber-Physical Systems
This research presents a method that integrates Large Language Models (LLMs), Retrieval-Augmented Generation (RAG), and Graph-based RAG (GraphRAG) for generating Design Structure Matrices (DSMs). The methods are tested on two distinct use cases: a power screwdriver and a CubeSat with known architectural references. The performance is evaluated on two key tasks: determining relationships between predefined components and the more complex challenge of identifying components and their subsequent relationships. Performance measurement involves assessing each element of the DSM as well as the overall architecture. Despite facing design and computational challenges, opportunities for automated DSM generation are identified, with all code made publicly available for reproducibility and further research.