基于知识图谱构建的煤矿机械检修系统  

Coal Mine Machinery Maintenance System Based on Knowledge Graph Construction

作　　者：崔亚仲 马力 李雄怀 高波 Cui Yazhong;Ma Li;Li Xionghuai;Gao Bo(CHN Energy Shendong Coal Group Co.,Ltd.,Yulin 719300,China)

机构地区：[1]国能神东煤炭集团有限责任公司,陕西榆林719300

出　　处：《煤矿机械》2024年第12期144-146,共3页Coal Mine Machinery

基　　金：国家重点研发计划项目(2017YFC0804300)。

摘　　要：为解决采煤机、转载机等井下设备检修存在的以人工为主、效率低、经验依赖等现实问题,通过构建煤机检修自学习知识图谱,形成数据采集、知识建设、知识融合、知识管理、知识应用五大模块,通过采集机械初始数据及运行日常监测数据,结合历史维修记录和高级专家经验,形成可解释机器学习模型及人机双向增强与反馈机制,并在设备运维领域进行应用,能够有效降低单独由机器或人决策而带来的系统运行风险,实现安全实时自检。另外,通过增加智能问答模块,可实现给定问题的再分解,通过知识库抽取匹配答案,自动检测其在时间与空间上的吻合度。最后将答案进行合并,以直观的方式展现最佳处理结果。研究成果可为煤矿智能化建设提供参考依据。In order to solve the practical problems such as manual work,low efficiency and experience dependence in the maintenance of underground equipment such as shearer and transfer,by constructing the self-learning knowledge graph of coal machine maintenance,five modules of data acquisition,knowledge construction,knowledge fusion,knowledge management and knowledge application were formed.By collecting the initial data of machinery and the daily monitoring data of operation,combined with historical maintenance records and senior expert experience,an interpretable machine learning model and a human-machine two-way enhancement and feedback mechanism were formed,and applied in the field of equipment operation and maintenance,which can effectively reduce the system operation risk caused by machine or human decision-making alone,and realize safe and real-time self-checking.In addition,by adding an intelligent question answering module,the given problem can be redecomposed,the matching answers are extracted through the knowledge base,and their coincidence in time and space is automatically detected.Finally,the answers are merged to show the best processing results in an intuitive way.The research results can provide reference for intelligent construction of coal mine.

关 键 词：知识图谱 煤机检修 智能化 数据模块 

分 类 号：TD67[矿业工程—矿山机电]