随着煤矿智能化技术快速发展，煤矿装备复杂化程度不断提高，装备维修资源不断丰富，传统的煤矿装备维护知识管理技术已经不能满足目前设备维护知识管理的需求。知识利用率低、互通性差、流失严重等问题逐步显现，因此亟需研究新的面向大规模煤矿装备维护资源的知识体系构建方法与知识管理应用技术。知识图谱是利用图模型描述客观世界事物间关联关系的技术方法，能够有效解决大规模数据下知识的动态挖掘与管理难题。因此本文重点研究煤矿装备维护知识图谱构建的关键技术，主要研究内容如下：

首先，基于煤矿装备维护领域没有统一基础知识体系的现状，本文通过建立煤矿装备维护领域本体模型（Coal Mine Equipment Maintenance Ontology，CMEMO），以有效解决该领域没有统一用于表示、整合与共享煤矿装备维护知识的难题，为煤矿装备维护知识图谱构建提供支持。

其次，针对传统的命名实体识别方法存在的识别效果差且特征提取不充分等问题，本文提出一种基于神经网络的煤矿装备维护命名实体识别模型(BERT-BiLSTM-CRF)，并将该模型应用到煤矿装备维护数据集进行验证。 

最后，通过需求分析与架构设计，采用Django应用框架搭建煤矿装备维护知识图谱系统，实现知识图谱系统各模块功能，通过实例对煤矿装备维护知识图谱系统的可靠运行进行验证。

综上所述，本文通过对煤矿装备维护知识图谱构建关键技术的研究与改进，所构建的煤矿装备维护知识图谱可实现对煤矿装备维护资源的有效组织与管理，并为煤矿装备维护提供决策支持。

With the rapid development of coal mine intelligence technology, the complexity of coal mine equipment continues to increase, and equipment maintenance resources continue to be enriched. Traditional coal mine equipment maintenance knowledge management technology can no longer meet the current needs of equipment maintenance knowledge management. Problems such as low utilization rate of knowledge, poor interoperability, and serious loss are gradually emerging, and it is urgent to study new knowledge construction and management application technologies for large-scale coal mine equipment maintenance resources. Knowledge graph is a method and technology that uses graph models to describe the relationship between things in the objective world, which can effectively solve the problem of dynamic knowledge mining and management under large-scale data. Therefore, this paper focuses on the use of knowledge graph technology to establish a knowledge graph system for coal mine equipment maintenance. The main research contents are as follows:

First of all, based on the current situation that there is no unified basic knowledge system in the field of coal mine equipment maintenance, this paper establishes a coal mine equipment maintenance field ontology (CMEMO) to effectively solve the problem that this field is not uniformly used to express, integrate and share coal mine equipment maintenance. The problem of knowledge provides support for the construction of knowledge graph of coal mine equipment maintenance.

Then, aiming at the problems of traditional named entity recognition methods such as poor recognition effect and insufficient feature extraction. This paper proposes to apply the BERT-BiLSTM-CRF model to named entity recognition for coal mine equipment maintenance, and applies the model to the coal mine equipment maintenance data set for verification. The experimental results show that under the same data set, the model entity recognition The effect is more advanced than other models.

Finally, through demand analysis and architecture design, combined with the constructed coal mine equipment maintenance domain ontology model, the BERT-BiLSTM-CRF model is used to complete the coal mine equipment maintenance entity identification, and the Django application framework is used to build a coal mine equipment maintenance knowledge graph system to realize the functions of each module of the knowledge graph system. The functions of each module of the system are verified through examples to verify the reliability and practicability of the coal mine equipment maintenance system.

In summary, the coal mine equipment maintenance knowledge graph proposed and constructed in this paper can realize the effective organization and management of coal mine equipment maintenance knowledge, and provide decision support for coal mine equipment maintenance.

[1]王国法,任怀伟,庞义辉,等. 煤矿智能化(初级阶段)技术体系研究与工程进展[J]. 煤炭科学技术,2020,48(07):1-27.

[2]王国法,刘峰,庞义辉,等. 煤矿智能化——煤炭工业高质量发展的核心技术支撑[J]. 煤炭学报, 2019,44(02):349-357.

[3]任怀伟,王国法,赵国瑞,等. 智慧煤矿信息逻辑模型及开采系统决策控制方法[J]. 煤炭学报,2019,44(09):2923-2935.

[4]李梅,杨帅伟,孙振明,等. 智慧矿山框架与发展前景研究[J]. 煤炭科学技术,2017,45(01):121-128+134.

[5]韩茜. 智慧矿山信息化标准化系统关键问题研究[D]. 中国矿业大学(北京),2016.

[6]邓仕拓. 露天煤矿大型设备维修信息管理平台的设计与实现[D]. 内蒙古大学,2019.

[7]丁恩杰,廖玉波,张雷,等. 煤矿信息化建设回顾与展望[J]. 工矿自动化, 2020,46(07):5-11.

[8]张婷婷,肖伟,都明基,等. 煤矿企业设备管理信息系统的研究与设计[J]. 中国煤炭,2018,44(07):85-87.

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

[10]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/OL]. 机械工程学报:1-24[2021-03-13]

[11]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.

[12]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.

[13]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.

[14]王国法,刘峰,孟祥军,等. 煤矿智能化(初级阶段)研究与实践[J]. 煤炭科学技术,2019,47(8):1-36

[15] 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.

[16]谭章禄,马营营,袁慧. 煤炭大数据平台建设的关键技术及管理协同架构[J]. 工矿自动化,2018,44(6):16-20.

[17]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.

[18]张轩. 基于WebGIS的煤矿设备管理系统研发[D]. 西安科技大学,2014.

[19]许欢,李敬兆. 主动感知的煤矿井下设备全生命周期自动化管理系统[J]. 佳木斯大学学报(自然科学版),2018,36(05):696-698+761.

[20]陈骋,祝钊. 煤矿设备管理系统设计与开发[J]. 煤炭工程,2015,47(06):116-118+122.

[21]秦桂英,刘宇熹,马海峰. NET技术在煤矿采煤设备维修管理中的应用[J]. 煤矿机械,2013,34(10):278-280.

[22]卢爱芹,陈维华. 基于Cordys的煤矿设备维护管理系统设计[J]. 煤炭技术,2012,31(12):256-257.

[23]乔杰,曹现刚. 基于B/S模式的煤矿设备维护管理系统研究[J]. 煤矿机械,2012,33(09):283-284.

[24]Berners-Lee T, Hendler J, Lassila O. The semantic web[J]. Scientific american, 2001, 284(5): 34-43.

[25]Shadbolt N, Berners-Lee T, Hall W. The semantic web revisited[J]. IEEE intelligent systems, 2006, 21(3): 96-101.

[26] Banko M, Cafarella M J, Soderland S, et al. Open information extraction for the web[C]//IJCAI. 2007, 7: 2670-2676.

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

[28] Pasca M, Lin D, Bigham J, et al. Organizing and searching the world wide web of facts-stepone: the one-million fact extraction challenge[C]//AAAI. 2006, 6: 1400-1405.

[29]D. B. Lenat and R. V. Guha. Building large knowledge-based systems; representation and inference in the Cyc project[J]. Addison-Wesley Longman Publishing Co., Inc., 1989.

[30]Marti A . WordNet : An Electronic Lexical Database[J]. Automated Discovery of Wordnet Relations, 1998, 5.

[31]Speer R, Havasi C. Representing general relational knowledge in conceptnet 5[C]//LREC. 2012: 3679-3686.

[32]百度知心. 百度百科[EB/OL].http://baike.baidu.com/view/10972128.html

[33] 搜狗知立方. 搜狗百科[EB/OL]. http://baike.sogou.com/

[34]S.Gallagher. How Google and Microsoft taught search to understand the Web[EB/OL]. http://arstechnica.com/information-technology/2012/06/inside-the-architecture-of-googles-knowledge-grapb-and-microsofls -satori/

[35]李景. 领域本体的构建方法与应用研究[D]. 中国农业科学院,2009.

[36]Hoang Long Nguyen, Dang Thinh Vu, Jason J. Jung,Knowledge graph fusion for smart systems: A Survey[J].Information Fusion,Volume 61, 2020.

[37] Philipp Kestel, Patricia Kügler, Christoph Zirngibl, et al,Ontology-based approach for the provision of simulation knowledge acquired by Data and Text Mining processes[J]. Advanced Engineering Informatics,Volume 39, 2019

[38]Xiaojun Chen, Shengbin Jia, Yang Xiang,A review: Knowledge reasoning over knowledge graph[J]. Expert Systems with Applications,Volume 141, 2020.

[39]K.W. Chau. An ontology-based knowledge management system for flow and water quality modeling[J]. Advances in Engineering Software, 2006,38(3).

[40]陈刚,陆汝钤,金芝. 基于领域知识重用的虚拟领域本体构造[J]. 软件学报,2003(03):350-355.

[41]Dong Yang,Rui Miao,Hongwei Wu, et al. Product configuration knowledge modeling using ontology web language[J]. Expert Systems With Applications,2008,36(3).

[42]王念滨,徐晓飞,王刚,等. 一种CIMS知识集成系统的研究与设计[J]. 计算机集成制造系统-CIMS,2000(05):29-34.

[43]代风,翟翔,施国强,等. 面向航天产品研制的知识网络本体建模方法[J]. 浙江大学学报(工学版),2018,52(10):2023-2034.

[44]剡昌锋,栗宇,王慧滨,等. 基于本体的汽轮发电机组故障诊断知识建模[J]. 兰州理工大学学报, 2020,46(05):41-48.

[45]王宁,葛瑞芳,苑春法,等. 中文金融新闻中公司名的识别[J]. 中文信息学报,2002(02):1-6.

[46]周昆. 基于规则的命名实体识别研究[D]. 合肥工业大学,2010.

[47]闫丹辉,毕玉德. 基于规则的越南语命名实体识别研究[J]. 中文信息学报,2014,28(05):198-205+214.

[48]姜伟. 基于规则的中文人名识别与抽取关键技术研究[J]. 科技创新导报,2012(28):65-66.

[49]木合塔尔·艾尔肯,艾斯卡尔·艾木都拉,地里木拉提·吐尔逊.基于规则的维吾尔地名识别[J]. 通信技术,2013,46(07):103-105.

[50]徐梓豪. 基于统计模型的中文命名实体识别方法研究及应用[D]. 北京化工大学,2017.

[51]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.

[52]孙僖. 垂直领域知识图谱构建的关键技术研究[D]. 北京邮电大学,2019.

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

[54]栗伟,赵大哲,李博,彭新茗,刘积仁.CRF与规则相结合的医学病历实体识别[J].计算机应用研究,2015,32(04):1082-1086.

[55]王国昱. 基于深度学习的中文命名实体识别研究[D]. 北京工业大学,2015.

[56]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.

[57]Luo Ling,Yang Zhihao,Yang Pei, et al. An attention-based BiLSTM-CRF approach to document-level chemical named entity recognition.[J]. Bioinformatics (Oxford, England), 2018,34(8).

[58]谢腾,杨俊安,刘辉. 基于BERT-BiLSTM-CRF模型的中文实体识别[J]. 计算机系统应用,2020,29(07):48-55.

[59]Dominik Tomaszuk,David Hyland-Wood. RDF 1.1: Knowledge Representation and Data Integration Language for the Web[J]. Symmetry, 2020,12(1).

[60]柏玉. 面向网络数据的信息抽取研究与应用[D]. 西南交通大学,2015.

[61]邓志鸿,唐世渭,张铭,等. Ontology研究综述[J]. 北京大学学报(自然科学版),2002(05):730-738.

[62]曹现刚,张梦园,雷卓,等. 煤矿装备维护知识图谱构建及应用[J]. 工矿自动化,2021,47(03):41-45.

[63]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.

[64]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.

[65]岳丽欣,刘文云. 国内外领域本体构建方法的比较研究[J]. 情报理论与实践, 2016,39(08):119-125.

[66]刘宇松. 本体构建方法和开发工具研究[J]. 现代情报,2009,29(09):17-24.

[67]Rabuzin Kornelije,Kudelić Robert. Exploring graph databases[J]. Politehnika: Časopis za tehnički odgoji obrazovanje,2020,4(2).

[68]叶帅. 基于Neo4j的煤矿领域知识图谱构建及查询方法研究[D]. 中国矿业大学,2019.

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

[70]Cross J, Huang L. Incremental parsing with minimal features using bi-directional LSTM[J]. arXiv preprint arXiv:1606.06406, 2016.

[71]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.

[72]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.

[73]GB 3836.13-2013. 爆炸性环境第13部分:设备的修理、检修、修复和改造[S]. 中华人民共和国国家质量监督检验检疫总局, 2013.12.17.

[74]GB/T 15663.10-2008. 煤矿科技术语第10部分:采掘机械[S]. 中华人民共和国国家质量监督检验检疫总局, 2008.07.

[75]Duque-Ramos A, Boeker M, Jansen L,et al. Evaluating the Good Ontology, Design Guideline (GoodOD) with the Ontology Quality Requirements and Evaluation Method and Metrics (OQuaRE)[J]. PLoS ONE 2014,9,1-14.

[76]Gu R, Hu W, Huang Y. Rainbow: a distributed and hierarchical RDF triple store with dynamic scalability[C]//2014 IEEE International Conference on Big Data (Big Data). IEEE, 2014: 561-566.

[77]Partner J, Vukotic A, Watt N. Neo4j in Action[J]. Pearson Schweiz Ag, 2014.

[78]张青青. 领域本体RDF图数据存储与查询方法研究[D]. 中国民航大学,2017.

[79]王鑫,邹磊,王朝坤,等. 知识图谱数据管理研究综述[J]. 软件学报,2019,30(07):2139-2174.