兰州理工大学学报 ›› 2021, Vol. 47 ›› Issue (5): 30-37.

• 机械工程与动力工程 • 上一篇    下一篇

基于本体的汽轮机多源异构知识建模与融合

剡昌锋*1, 张永明1,2, 艾科勇1, 栗宇3, 吴黎晓1   

  1. 1.兰州理工大学 机电工程学院, 甘肃 兰州 730050;
    2.安徽容知日新科技股份有限公司, 安徽 合肥 230031;
    3.大秦铁路股份有限公司, 山西 太原 030045
  • 收稿日期:2019-12-19 出版日期:2021-10-28 发布日期:2021-11-18
  • 通讯作者: 剡昌锋(1974-),男,甘肃平凉人,博士,研究员,博导.Email:changf_yan@163.com
  • 基金资助:
    国家自然科学基金(51765034,51165018)

Ontology-based heterogeneous knowledge modeling and fusion of steam turbine

YAN Chang-feng1, ZHANG Yong-ming1,2, AI Ke-yong1, LI Yu3, WU Li-xiao1   

  1. 1. School of Mechnical and Electrical Engineering, Lanzhou Univ. of Tech., Lanzhou 730050, China;
    2. Anhui Ronds Science and Technology Co. Ltd, Hefei 230031, China;
    3. Daqin Railway Co. Ltd, Taiyuan 030045, China
  • Received:2019-12-19 Online:2021-10-28 Published:2021-11-18

摘要: 针对目前汽轮发电机组设备维修与故障诊断的知识分散在不同电厂内,普遍存在多源异构、共享困难以及形成信息孤岛等问题,结合全局本体与局部本体建模的方法,分析了汽轮发电机组的结构特性,借助Protégé_4.3建立了汽轮发电机组的全局本体模型与局部本体模型,设计了全局本体与局部本体的映射关系算法,实现了汽轮发电机组多源异构的知识融合与多源知识的检索.采用SQI机械故障模拟实验台对汽轮发电机组产生的故障进行模拟.通过模拟转子质量偏心和转子不平衡,结果表明两者有98%的相似度,即两个不同概念的故障有相同的故障源,说明基于本体的汽轮发电机组多源异构知识模型与融合方法是有效、可行的.

关键词: 汽轮发电机组, 知识建模, 全局本体, 局部本体, 映射关系

Abstract: In view of the problems that the knowledge of steam turbine generator equipment maintenance and fault diagnosis is scattered in different power plant, and there are widespread problems such as multi-source heterogeneity, sharing difficulties and the formation of information islands, the structural characteristics of steam turbine generator equipment are analyzed based on the methods of global ontology and local ontology modeling. The global ontology model and local ontology model of the steam turbine generator sets are established by means of Protégé_4.3. The mapping relationship between global ontology and local ontology is designed. The multi-source heterogeneous knowledge fusion and multi-source knowledge retrieval of turbo generator sets are realized. The SQI mechanical fault simulation test rig is used to simulate the faults generated by the steam turbine generator sets. The results show that they have 98% similarity, two different concept faults have the same fault source, which shows that the multi-source heterogeneous knowledge model and fusion method based on ontology is effective and feasible through the simulation of rotor mass eccentricity and rotor imbalance.

Key words: steam turbine generator sets, knowledge modeling, global ontology, local ontology, mapping relationship

中图分类号: