设备维护是制造业适应全球化进程的重要保障。设备维护知识对设备维护的发展具有重要的指导意义，但是设备维护知识来源广泛、数据量庞大、信息要素及结构复杂等因素使得知识的可用性较差。促进设备维护知识多源共享，挖掘设备维护知识内在关系，是促进设备维护发展的重要研究内容。知识图谱是将人类知识结构化形成的知识系统，也是人工智能研究和智能信息服务的核心技术，由于其具有赋予智能体深度理解、精准查询与逻辑推理等能力，被广泛的应用于各类垂直领域。构建设备维护知识图谱作为融合了人工智能的设备维护知识管理方式，能够整合设备维护领域中分散、关系复杂的数据信息，对设备维护知识进行高效管理和利用。

设备维护知识是构建领域知识图谱的核心资源，本文基于构建知识图谱的目的，考虑领域知识的专业性，首先分析本文研究的设备维护领域知识，设计了设备维护领域本体构建方法，并利用Protégé软件对构建的领域本体进行存储。

其次对设备维护实体类别进行了定义，并针对设备维护命名实体识别中存在的一词多义、嵌套实体等难点，设计了一种基于BERT-BiLSTM-CRF的设备维护命名实体识别模型。

然后研究了设备维护知识实体间的关系，基于实体识别结果，设计了一种基于R-BERT的设备维护实体关系抽取模型。

接着研究了设备维护知识存储技术，设计了映射规则并利用Neo4j图数据库进行知识存储，完成设备维护领域知识图谱构建。

再然后研究并完成了设备维护知识问答对设备维护知识图谱进行应用。最后基于设备维护知识图谱，设计了利用Django框架搭建前后端分离的设备维护知识图谱，并完善知识图谱系统各个模块功能，完成知识图谱系统可视化管理。

Equipment maintenance is an important guarantee for manufacturing industry to adapt to the process of globalization. Equipment maintenance knowledge has important guiding significance for the development of equipment maintenance, but the availability of equipment maintenance knowledge is poor due to such factors as wide sources, large amount of data, complex information elements and structure. It is an important research content to promote the development of equipment maintenance to promote multi-source sharing of equipment maintenance knowledge and to explore the internal relationship of equipment maintenance knowledge. Knowledge graph is a knowledge system that forms the structure of human knowledge, and it is also the core technology of artificial intelligence research and intelligent information service. Because of its ability to endue agents with deep understanding, accurate query and logical reasoning, it is widely used in various vertical fields. As a management method of equipment maintenance knowledge integrated with artificial intelligence, the construction of equipment maintenance knowledge map can integrate scattered and complex data information in the field of equipment maintenance and efficiently manage and utilize equipment maintenance knowledge.
Equipment maintenance knowledge is the core resource for constructing domain knowledge graph. Based on the purpose of constructing knowledge graph and considering the specialty of domain knowledge, this paper firstly analyzes the equipment maintenance domain knowledge studied in this paper, designs the equipment maintenance domain ontology construction method, and uses Protege software to store the constructed domain ontology.
Secondly, the classification of equipment maintenance entities is defined, and aiming at the difficulties of polysemous and nested entities in equipment maintenance named entity recognition, a device maintenance named entity recognition model based on Bert-BilSTM-CRF is designed.
Then, the relationship between entities of equipment maintenance knowledge is studied. Based on the result of entity recognition, an entity relationship extraction model of equipment maintenance based on R-Bert is designed.
Then, the device maintenance knowledge storage technology is studied, mapping rules are designed and knowledge is stored by using Neo4j graph database to complete the construction of device maintenance domain knowledge map.
Then research and complete the equipment maintenance knowledge q&A equipment maintenance knowledge graph application. Finally, based on the device maintenance knowledge graph, Django framework was designed to build the device maintenance knowledge graph separated from the front and back ends, and the functions of each module of the knowledge graph system were improved to complete the visual management of the knowledge graph system.

[1]AMIT S. Introducing the knowledge graph[R]. America:Official Blog of Google, 2012.

[2]张栋豪，刘振宇，郏维强，等. 知识图谱在智能制造领域的研究现状及其应用前景综述[J]. 机械工程学报，2021, 57(05): 90-113.

[3]刘烨宸，李华昱. 领域知识图谱研究综述[J]. 计算机系统应用，2020, 29(06): 1-12.

[4]徐增林，盛泳潘，贺丽荣，等. 知识图谱技术综述[J]. 电子科技大学学报，2016, 45(04): 589-606.

[5]刘峤,李杨,段宏,等. 知识图谱构建技术综述[J]. 计算机研究与发展，2016, 53(03): 582-600.

[6]朱海华. 客户需求驱动的产品服务系统在航空MRO领域的研究[D]. 南京：南京理工大学，2013.

[7]李小涛，胡晓惠，李斌全.基于两层元数据与本体的异构数据共享技术[J]. 北京航空航天大学学报，2015, 41(08): 1476-1484.

[8]陈飞. 基于维修BOM的复杂装备MRO服务管理系统的研究[D]. 浙江：浙江大学，2012.

[9]杨珺婷. 基于BOM映射的机车检修计划方法研究[D]. 大连：大连理工大学， 2021

[10]程曜安，张力，刘英博，等. 大型复杂装备MRO系统解决方案[J]. 计算机集成制造系统，2010, 16(10): 2026-2037.

[11]李鹏. 军医大学科研仪器设备知识地图构建的研究[D]. 重庆：第三军医大学，2004.

[12]刘朝晖. 基于语义网的设备知识管理[D]. 武汉：武汉科技大学，2011.

[13]曹现刚，罗璇，张鑫媛，等. 煤矿机电设备运行状态大数据管理平台设计[J]. 煤炭工程，2020, 52(2): 22-26.

[14]段欣宇. 面向物联网监控的煤矿设备异构数据集成与分布式存储研究[D]. 西安：西安科技大学，2021.

[15]魏顺平. 基于术语部件的领域本体自动构建方法研究—以教育技术学领域本体构建为例[J]. 电化教育研究, 2013, 34(5): 6.

[16]冯瑶, 冯锡炜. 基于本体的教育资源推理查询原型系统设计与实现[J]. 计算机应用与软件, 2016, 33(10): 6.

[17]王茜. 基于文本挖掘的动态本体构建方法研究[D]. 北京：中国农业大学，2007.

[18]Kafkas S, Hoehndorf R. Ontology Based Text Mining of Gene-phenotype Associations: Application to Candidate Gene Prediction[J]. The Journal of Biological Databases and Curation 2019: 1-9.

[19]Kethavarapu UPK, Saraswathi，S.Concept Based Dynamic Ontology Creation for Job Recommendation System[J]. Procedia Computer Science 2016, (85): 915-921.

[20]付苓. 基于大数据的领域本体动态构建方法研究—养生领域本体构建为例[J]. 情报理论与实践，2018, 41 (1) : 135-138.

[21]桂冬冬， 王向前，李慧宗. 煤矿事故本体自动构建[J]. 工矿自动化，2019, 45 (9) : 75-79.

[22]Carlson A, Betteridge J, Kisiel B, et al. Toward an Architecture for Never-Ending Language Learning[C]//AAAI. 2010, 5: 3.

[23]李涓子，侯磊. 知识图谱研究综述[J]. 山西大学学报：自然科学版，2017, 40(3): 6.

[24]韩吉南，袁鸿翼，郑奎志，等. 面向航天装备大数据的知识图谱体系[J]. 国防科技，2021, 42(05): 40-46.

[25]李连营，彭楚骄，郭邦祁，等. 海洋地图可视化方法的知识图谱构建[J]. 测绘地理信息，2022, 47(01): 77-80.

[26]李哲，周斌，李文慧，等. 煤矿机电设备事故知识图谱构建及应用[J]. 工矿自动化，2022, 48(01): 109-112..

[27]张雨琪，李宗友，王映辉，等. 中医药知识图谱的构建与应用研究[J]. 世界中医药，2022, 17(04): 553-558.

[28]孙镇，王惠临. 命名实体识别研究进展综述[J]. 现代图书情报技术，2010(06): 42-47.

[29]孙茂松，黄昌宁，高海燕，等. 中文姓名的自动辨识[J]. 中文信息学报，1995, 9(2): 16-27.

[30]彭春艳，张晖，包玲玉，等. 基于条件随机域的生物命名实体识别[J]. 计算机工程，2009, 35(22): 197-199.

[31]乐娟，赵玺. 基于HMM的京剧机构命名实体识别算法[J]. 计算机工程，2013, 39(6): 266-271.

[32]蓝雁玲. 基于多模板HMM的中文命名实体识别[D]. 广州：华南理工大学，2011.

[33]Hammerton James. Named entity recognition with long short-term memory[J]. Proceedings of the 7th Conference on Natural Language Learning at HLT-NAACL. Stroudsburg, PA, USA. 2003. 172-175.

[34]LAMPLE G, BALLESTEROS M, SUBRAMANIAN S, et al. Neural architectures for named entity recognition[C]// Proceedings of the 2016 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. San Diego, CA, USA. 2016. 260-270.

[35]MA X, HOVY E. End-to-end sequence labeling via bi-directional LSTM-CNN-CRF[J]. In Proc, in: 54th Annual Meeting of the Association for Computational Linguistics, 2016, pp. 1064-1074.

[36]Huang WM, Hu DR, Deng ZR, et al. Named entity recognition for Chinese judgment documents based on BiLSTM and CRF[J]. EURASIP Journal on Image and Video Processing, 2020, 2020(1).

[37]Tsai T H , Wu S H , Lee C W , et al. Mencius: A Chinese Named Entity Recognizer Using the Maximum Entropy-based Hybrid Model[J]. 中文计算语言学期刊, 2004, 9(1): 65-81.

[38]Huang Z, Wei X, Kai Y. Bidirectional LSTM-CRF Models for Sequence Tagging[J]. Computer Science, 2015.

[39]DEVLIN J, CHANG M, LEE K, et al. Bert: Pre-training of deep bidirectional transformers for language understanding[J]. arXivpreprint arXiv: 1810.04805, 2018.

[40]LIU J, GAO L, GUO S J, et al. A hybrid deep-learning approach for complex biochemical named entity recognition[J]. Knowledge-Based Systems, 2021, 221.

[41]PANG N, QIAN L, LYU W, et al. Transfer Learning for Scientific Data Chain Extraction in Small Chemical Corpus with BERT-CRF Model[J]. arXiv, 2019.

[42]许力，李建华. 基于BERT和BiLSTM-CRF的生物医学命名实体识别[J]. 计算机工程与科学，2021, 43(10): 1873-1879.

[43]漆桂林，高桓，吴天星. 知识图谱研究进展[J]. 情报工程，2017, 3(1):22.

[44]Liu C Y, Sun W B, Chao W H, et al. Convolution Neural Network for Relation Extraction[C]// International Conference on Advanced Data Mining and Applications. Springer Berlin Heidelberg, 2013.

[45]Bekoulis G, Deleu J, Demeester T, et al. Joint entity recognition and relation extraction as amulti-head selection problem[J]. Expert Systems with Applications, 2018, 114: 34-45.

[46]周新宇，李培峰. 基于信息交互增强的事件时序关系分类方法[J]. 计算机科学，2020, 47(11): 244-249.

[47]万莹，孙连英，赵平，等. 基于信息增强BERT的关系分类[J]. 中文信息学报，2021, 35(03): 69-77.

[48]Weizenbaum, Joseph. Eliza-a Computer Program for the Study of Natural Language Communication Between Man and Machine[J]. Communications of the Acm, 1966,9(1): 36-45.

[49]王智悦，于清，王楠，等. 基于知识图谱的智能问答研究综述[J]. 计算机工程与应用，2020, 56(23): 1-11.

[50]Cui W，Xiao Y，Wang H，et al. KBQA：learning question answering over QA corpora and knowledge bases[J]. Proceedings of the VLDB Endowment, 2017, 10(5): 565-576.

[51]Abujabal A，Yahya M，Riedewald M，et al. Automated template generation for question answering over knowledge graphs[C]//the Web Conference，2017：1191-1200.

[52]Cocco R，Atzori M，Zaniolo C，et al.Machine learning of SPARQL templates for question answering over Linked Spending[C]//Workshops on Enabling Technologies：Infrastracture for Collaborative Enterprises，2019：156-161.

[53]Kwiatkowski T，Zettlemoyer L，Goldwater S，et al.Lexical generalization in CCG grammar induction for semantic parsing[C]//Conference on Empirical Methods in Natural Language Processing，2011：1512-1523.

[54]Berant J，Chou A K，Frostig R，et al. Semantic parsing on free base from question-answer pairs[C]//Conference on Empirical Methods in Natural Language Processing, 2013: 1533-1544.

[55]Reddy S, Tackstrom O, Collins M, et al. Transforming dependency structures to logical forms for semantic parsing[J]. Transactions of the Association for Computational Linguistics, 2016, 4(1): 127-140.

[56]Yih W, Chang M, He X, et al. Semantic parsing viastaged query graph generation：question answering with knowledge base[C]//International Joint Conference on Natural Language Processing, 2015: 1321-1331.

[57]刘园园，李劲华，赵俊莉. 基于语义解析的领域问答系统的设计与实现[J]. 计算机应用与软件，2021, 38(11): 42-48+97.

[58]Chen B，Sun L，Han X，et al. Sequence-to-action：end-to-end semantic graph generation for semantic parsing[C]//Meeting of the Association for Computational Linguistics, 2018: 766-777.

[59]Dong L, Lapata M. Coarse-to-fine decoding for neural semantic parsing[C]//Meeting of the Association for Computational Linguistics, 2018: 731-742.

[60]Lukovnikov D, Fischer A, Lehmann J, et al. Neural network-based question answering over knowledge graphs on word and character level[C]//the Web Conference, 2017: 1211-1220.

[61]陈晓军，向阳． 企业风险知识图谱的构建及应用[J]. 计算机科学，2020, 47(11): 237-243.

[62]田迎，单娅辉，王时绘． 基于知识图谱的抑郁症自动问答系统研究[J]. 湖北大学学报(自然科学版)，2020, 42(05): 587-591+596.

[63]黄东晋，梁景坤，李娜，等. 基于Bi-LSTM的动画电影智能问答系统[J]. 现代电影技术，2020(05): 30-35+41.

[64]翟林. 领域本体的半自动构建方法研究与实现[D]. 南京：东南大学，2005.

[65]李丽双，郭元凯. 基于CNN-BLSTM-CRF模型的生物医学命名实体识别[J]. 中文信息学报，2018, 32(01): 116-122.

[66]王向前，张宝隆，李慧宗. 本体研究综述[J]. 情报杂志，2016, 35(06): 163-170.

[67]姜亮，孙铁兵，屈福政. 基于本体的技术创新研究[J]. 中国机械工程，2008(15): 1853-1857.

[68]Samridha M, Akshar Chawla, Shagnik Roy, et al. Semantic Network[J]. International Journal of Engineering and Advanced Technology (IJEAT), 2020,10(1).

[69]周安美，于德介，刘坚，等. 基于本体和FTF的风力发电设备维护优化[J]. 中国机械工程，2014, 25(13): 1748-1754.

[70]刘奇林. 基于语义的工业生产设备预测性维护策略研究与实现[D]. 重庆：重庆邮电大学，2020.

[71]Devlin J, Chang MW, Lee K, et al. Bert: Pre-training of deep bidirectional transformers for language understanding[J]. arXiv preprint arXiv: 1810.04805, 2018.

[72]Vaswani A, Shazeer N, Parmar N, et al. Attention is all you need[C]//Advances in Neural Information Processing Systems 30.NewYork: Curran Associates,2017: 5998-6008.

[73]Felix A. Gers, Jü, rgen Schmidhuber,et al. Learning to Forget: Continual Prediction with LSTM[J]. Neural Computation,2000,12(10)

[74]Lafferty J, McCallum A, Pereira F C N. Conditional random fields: Probabilistic models for segmenting and labeling sequence data[C]// Proc. 18th International Conf. on Machine Learning, 2001, S. 282-289.