大型铸锻件热处理知识图谱构建及可溯源检索增强问答方法Construction of Knowledge Graph and Traceable Retrieval-Augmented QuestionAnswering Method for Heat Treatment of Large Castings and Forgings
徐恩献,王立,曹胜国,吴子健
摘要(Abstract):
大型铸锻件热处理工艺制定涉及大量分散的国家标准、行业标准、企业标准、工艺规程、操作手册、检验报告与记录卡等文档,长期存在格式异构、检索困难、知识分散、难传承以及问答结果溯源困难等问题。针对上述问题,提出了一种面向大型铸锻件热处理的领域知识图谱与可溯源检索增强问答方法。该方法以多格式文档统一接入为入口,对PDF、办公文档、扫描件与图片等进行解析、逐页视觉识别、清洗与分块,并行构建向量索引、关键词索引以及知识图谱;在此基础上自动抽取材料、工艺、设备、标准和缺陷等实体及其关系,构建“混合召回—交叉编码精排—逐点引用”的检索增强生成问答链路,将向量召回与关键词召回经倒数排名融合后进行精排,叠加资料可信度与来源权重,生成逐点标注来源且可跳转原文的可溯源答案;同时以业务域作为权限与数据隔离边界,防止越权资料进入检索上下文。工程应用表明,该方法能够将分散的热处理标准与工艺文档统一组织为可检索、可问答、可溯源的知识资产,提高工艺人员的资料检索与知识获取效率,为大型铸锻件热处理工艺研发的数字化、智能化转型提供支撑。
关键词(KeyWords): 大型铸锻件;热处理;工业知识库;知识图谱;检索增强生成;可溯源问答
基金项目(Foundation):
作者(Author): 徐恩献,王立,曹胜国,吴子健
DOI: 10.14147/j.cnki.51-1396/tg.2026.05.005
参考文献(References):
- [1]Dossett J L, Boyer H E. Practical Heat Treating[M]. 2nd ed. Materials Park:ASM International,2006.
- [2]Buchgeher G, Gabauer D, Martinez-Gil J, et al.Knowledge graphs in manufacturing and production:A systematic literature review[J]. IEEE Access,2021,9:55537-55554.
- [3]Lewis P,Perez E,Piktus A,et al. Retrieval-augmented generation for knowledge-intensive NLP tasks[C]//Advances in Neural Information Processing Systems 33. Red Hook:Curran Associates, 2020:9459-9474.
- [4]Xu Y,Li M,Cui L,et al. Layout LM:Pre-training of text and layout for document image understanding[C]//Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery&Data Mining. New York:ACM Press,2020:1192-1200.
- [5]Du Y,Li C,Guo R,et al. PP-OCR:A practical ultra lightweight OCR system[EB/OL].(2020-09-21)[2026-09-15]. https://arxiv. org/abs/2009.09941.
- [6]Hogan A,Blomqvist E,Cochez M,et al. Knowledge graphs[J]. ACM Computing Surveys,2021,54(4):1-37.
- [7]Karpukhin V,Oguz B,Min S,et al. Dense passage retrieval for open-domain question answering[C]//Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing. Stroudsburg:Association for Computational Linguistics, 2020:6769-6781.
- [8]Reimers N, Gurevych I. Sentence-Bert:Sentence embeddings using Siamese BERT-networks[C]//Proceedings of EMNLP-IJCNLP 2019. Stroudsburg:Association for Computational Linguistics,2019:3980-3990.
- [9]Robertson S,Zaragoza H. The probabilistic relevance framework:BM25 and beyond[J]. Foundations and Trends in Information Retrieval,2009,4(1-2):1-174.
- [10]Cormack G V,Clarke C L A,Buettcher S. Reciprocal rank fusion outperforms Condorcet and individual rank learning methods[C]//Proceedings of the 32nd International ACM SIGIR Conference on Research and Development in Information Retrieval. New York:ACM Press,2009:758-759.
- [11]Nogueira R,Cho K. Passage re-ranking with BERT[EB/OL].(2019-01-11)[2026-09-15]. https://arxiv. org/abs/1901. 04085.
- [12]Lin J,Ma X,Lin S C,et al. Pyserini:A Python toolkit for reproducible information retrieval research with sparse and dense representations[C]//Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval. New York:ACM Press,2021:2356-2362.
- [13]Shuster K,Poff S,Chen M,et al. Retrieval augmentation reduces hallucination in conversation[C]//Findings of the Association for Computational Linguistics:EMNLP 2021. Stroudsburg:Association for Computational Linguistics,2021:3784-3803.
- [14]Asai A,Wu Z,Wang Y,et al. Self-RAG:Learning to retrieve,generate,and critique through self-reflection[C]//The Twelfth International Conference on Learning Representations. Addis Ababa:ICLR,2024.
- [15]Liu N F,Lin K,Hewitt J,et al. Lost in the middle:How language models use long contexts[J].Transactions of the Association for Computational Linguistics,2024,12:157-173.