中国电力 ›› 2024, Vol. 57 ›› Issue (11): 78-87.DOI: 10.11930/j.issn.1004-9649.202406047

• 电网 • 上一篇    下一篇

标准数字化应用框架设计及其设备侧领域实践

贺馨仪1(), 董明1(), 孙歆2, 颜拥2, 姚影2   

  1. 1. 西安交通大学 电力设备电气绝缘国家重点实验室,陕西 西安 710049
    2. 国网浙江省电力有限公司电力科学研究院,浙江 杭州 310007
  • 收稿日期:2024-06-11 出版日期:2024-11-28 发布日期:2024-11-27
  • 作者简介:贺馨仪(1998—),女,通信作者,硕士研究生,从事电力大数据与电力标准数字化研究,E-mail:zhufeng703@stu.xjtu.edu.cn
    董明(1977—),男,教授,从事电气设备状态监测及人工智能应用研究,E-mail: dongming@xjtu.edu.cn
  • 基金资助:
    国家电网有限公司科技项目(5700-202241437A-2-0-ZN)。

The Design of Standard Digital Application Framework and its Field Practice on Device Side

Xinyi HE1(), Ming DONG1(), Xin SUN2, Yong YAN2, Ying YAO2   

  1. 1. The State Key Laboratory of Electrical Insulation for Power Equipment, Xi'an Jiaotong University, Xi'an 710049, China
    2. State Grid Zhejiang Electric Power Company Electric Power Research Institute, Hangzhou 310007, China
  • Received:2024-06-11 Online:2024-11-28 Published:2024-11-27
  • Supported by:
    This work is supported by Science and Technology Project of SGCC (No.5700-202241437A-2-0-ZN).

摘要:

为了满足对电力标准文献的深度挖掘和知识规则构建的需求,提出了一种电力标准数字化方案,该方案以条件事理图谱为基础,为电力行业的高效、智能化发展提供了新的解决方案。首先为实现电力标准文献的精细化知识抽取,结合了规则和语义角色标注技术,提取出关键知识三元组,在知识融合阶段,采用词语相似度方法整合知识,提高图谱准确性;其次,建立了针对电力领域的改进标准知识图谱,更全面地表达了电力标准中的知识和规则;最后,结合Cypher查询语言和产生式规则,实现电力设备的辅助知识问答和状态评估任务,大大增强了图谱的实用性。

关键词: 条件事理图谱, 电力标准数字化, 知识三元组抽取

Abstract:

In order to meet the demand for deep mining and knowledge rule construction of electric power standard literature, this paper proposed a novel electric power standard digitization technology, which is based on conditional event graph and provided a new solution for the efficient and intelligent development of the electric power industry. Firstly, in order to realize the refined knowledge extraction of electric power standard documents, this paper combined the rule and semantic role annotation techniques to extract the key knowledge triad, and in the knowledge fusion stage, the word similarity method was used to integrate the knowledge and improve the accuracy of the atlas. Secondly, this paper established an improved standard knowledge graph for the electric power domain, which expresses the knowledge and rules in electric power standards more comprehensively. Finally, this paper combined Cypher query language and generative rules to realize the auxiliary knowledge quiz and condition assessment tasks of electric power equipment, which greatly enhances the utility of the graph.

Key words: condition event graph, digital power standards, knowledge triplet extraction