中国机械工程 ›› 2026, Vol. 37 ›› Issue (4): 967-976.DOI: 10.3969/j.issn.1004-132X.2026.04.021

• 再制造与退役产品资源化技术 • 上一篇    下一篇

基于图检索增强生成的报废动力电池拆解序列规划

王航1(), 鄢威1(), 张绪美1, 朱硕2, 江志刚2, 朱泽睿1   

  1. 1.武汉科技大学汽车与交通工程学院, 武汉, 430065
    2.武汉科技大学机械传动与制造工程湖北省重点实验室, 武汉, 430081
  • 收稿日期:2025-04-14 出版日期:2026-04-25 发布日期:2026-05-11
  • 通讯作者: 鄢威
  • 作者简介:王航,男,2001年生,硕士研究生。研究方向为绿色制造。E-mail: 204362470@qq.com
    鄢威*(通信作者),男,1981年生,教授、博士研究生导师。研究方向为绿色制造与再制造、智慧物流。E-mail: yanwei81@wust.edu.cn
  • 基金资助:
    国家自然科学基金(52075396);武汉科技大学“十四五”湖北省优势特色学科项目(2023B0405)

GraphRAG-based Disassembly Sequences Planning for End-of-life Power Batteries

WANG Hang1(), YAN Wei1(), ZHANG Xumei1, ZHU Shuo2, JIANG Zhigang2, ZHU Zerui1   

  1. 1.School of Automobile and Traffic Engineering,Wuhan University of Science and Technology,Wuhan,430065
    2.Hubei Key Laboratory of Mechanical Transmission and Manufacturing Engineering,Wuhan University of Science and Technology,Wuhan,430081
  • Received:2025-04-14 Online:2026-04-25 Published:2026-05-11
  • Contact: YAN Wei

摘要:

针对报废动力电池拆解知识难以复用所导致的序列生成效率低下问题,融合知识图谱(KG)在结构化知识表征与大语言模型在语义推理方面的互补优势,提出了一种基于图检索增强生成的报废动力电池拆解序列规划方法。首先提出基于KG标签匹配的子图生成方法,通过Cypher查询语言形成待拆解特定型号电池的专属拆解序列子图;然后建立基于混合检索和重排序的拆解知识粗粒度检索机制,以实现对目标电池组件的精确定位;最后,构建基于层级约束关系多跳推理的拆解知识细粒度检索模式,通过提取与组件关联的拆解序列信息,利用大语言模型实现拆解序列的智能生成。实验结果显示,所提方法在5种主流动力电池拆解序列生成上实现93.9%的准确率,展现出优异的可行性和有效性。

关键词: 拆解序列规划, 图检索增强生成, 知识图谱, 大语言模型, 报废动力电池

Abstract:

To address the challenges of low-efficiency disassembly sequence generation caused by the lack of knowledge reusability of end-of-life power batteries, a disassembly sequence planning method was proposed based on GraphRAG, by integrating the complementary strengths of KGs in structured knowledge representation and LLMs in semantic reasoning. First, a KG label-matching-based subgraph generation method was proposed, utilizing the Cypher query language to form custom disassembly sequence subgraphs for specific battery models. Second, a coarse-grained disassembly knowledge retrieval mechanism employing hybrid retrieval and re-ranking was established to locate target battery components precisely. Finally, a fine-grained retrieval model of disassembly knowledge was constructed based on multi-hop reasoning of hierarchical constraint relations. By extracting the disassembly sequence information associated with the components, the intelligent generation of disassembly sequences was achieved using the large language model. The experimental results indicate that the proposed method achieves an accuracy of 93.9% in disassembly sequence generation across five mainstream power batteries, demonstrating the excellent feasibility and effectiveness.

Key words: disassembly sequence planning, graph retrieval-augmented generation(GraphRAG), knowledge graph(KG), large language model(LLM), end-of-life power battery

中图分类号: