基于知识图谱的故障预测知识结构与热点迁徙研究
作者:
作者单位:

南京理工大学

作者简介:

通讯作者:

中图分类号:

TP18

基金项目:


Research on Knowledge Structure and Hotspots Migration of Prognostics based on Knowledge Map
Author:
Affiliation:

Nanjing University of Science and Technology

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    故障预测研究因应用对象广泛、技术理论先进和实用价值高而备受关注。本文从“文献追踪”视角,挖掘故障预测的知识结构、分布脉络和研究热点,这是对故障预测综述类研究的一个新尝试。研究发现:(1)知识结构方面,故障预测与状态监测、健康管理间存在较强的耦合关联性。模型驱动、知识驱动、统计驱动、概率推理方法、机器学习和深度学习是故障预测关键主要技术类别。(2)热点迁徙方面,故障预测研究主要经历了理论奠基期、内涵延伸期、技术涌现期和方法融合期四个阶段。本文归纳了阶段成果、面临困境及发展贡献,并指出阶段间衔接关系。探明了故障预测理论发展轨迹和实践成效,并实现故障预测领域的阶跃式发展提供明确的方向,即提高大数据收集质量、在线预测和跨工况的迁移学习。

    Abstract:

    Prognostics research has attracted much attention due to its wide range of application objects, advanced technical theory and high practical value. In this paper, from the perspective of "literature tracking", we try to mining and analyze the knowledge structure, distribution context and research hotspots of prognostics, which would be a new attempt to the research of prognostics review. The results show that: (1) in terms of knowledge structure, prognostics has strong coupling correlation with condition monitoring and health management. Model driven, knowledge driven, statistical driven, probabilistic reasoning methods, machine learning and deep learning are the key technical categories of prognostics. (2) In terms of hotspots migration, the research on prognostics mainly includes four periods: theoretical foundation period, connotation extension period, technology emergence period and method integration period. This paper summarizes the current achievements, difficulties faced and development contributions, clarifies the development track and practical effect of prognostics theory, and provides a clear direction for the step development of prognostics in the future, which quality improvement for collected big data, online prediction and migration learning will be the main trends in future development.

    参考文献
    相似文献
    引证文献
引用本文
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2020-10-12
  • 最后修改日期:2021-03-08
  • 录用日期:2021-03-16
  • 在线发布日期:
  • 出版日期: