基于知识图谱的电力调度自动化作业工作票生成方法  

A knowledge graph-based work order generation method for power dispatch automation operations

作　　者：李泽科 陈书里 陈斌 陈荃韡 杨旭 许贻杰 刘延华[4] LI Zeke;CHEN Shuli;CHEN Bin;CHEN Quanwei;YANG Xu;XU Yijie;LIU Yanhua(State Grid Fujian Electric Power Co.,Ltd.,Fuzhou,Fujian 350012,China;Ultra High Voltage Branch,State Grid Fujian Electric Power Co.,Ltd.,Fuzhou,Fujian 350013,China;State Grid Info-Telecom Great Power Science and Technology Co.,Ltd.,Fuzhou,Fujian 350003,China;College of Computer and Data Science,Fuzhou University,Fuzhou,Fujian 350108,China)

机构地区：[1]国网福建省电力有限公司,福建福州350012 [2]国网福建省电力有限公司超高压分公司,福建福州350013 [3]国网信通亿力科技有限责任公司,福建福州350003 [4]福州大学计算机与大数据学院,福建福州350108

出　　处：《福州大学学报(自然科学版)》2025年第2期135-143,共9页Journal of Fuzhou University(Natural Science Edition)

基　　金：福建省自然科学基金资助项目(2021J01625);国网福建省电力有限公司科技项目(521300230007)。

摘　　要：针对电力调度自动化作业工作票自动生成问题,采用基于上下文信息的多源数据处理方法,将作业文档转化为结构化记录和非结构化文本;通过设计融合类别信息的实体关系联合抽取模型,结合双向编码器表示等技术,对作业文档中的实体和关系进行抽取,构建领域知识图谱;进而基于该知识图谱,建立工作票自动生成算法模型,用于生成工作票中的工作内容、作业设备及安全措施.实验结果表明,与基线模型对比,实体关系联合抽取的精确度、召回率以及准确率和召回率的调和平均数(F 1分数)分别提升了31.8%、48.7%和46.5%.自动生成的工作票在准确性和完整性上优于人工编制,能够满足实际需求.Addressing the automatic generation of work tickets for power dispatch automation operations,a multi-source data processing method based on contextual information is adopted to transform operation documents into structured records and unstructured text.By designing a category-aware joint entity relation extraction model that integrates category information,combined with bidirectional encoder representations and other techniques,entities and relationships within the operation documents are extracted to construct a domain-specific knowledge graph.Based on this knowledge graph,an algorithm model for the automatic generation of work tickets is established,which generates the content of work tickets including job descriptions,involved equipment,and safety measures.Experimental results show that compared to baseline models,the precision,recall,and the harmonic mean of precision and recall(F 1 score)of entity relation joint extraction have increased by 31.8%,48.7%and 46.5%,respectively.Case analysis demonstrates that automatically generated work tickets surpass manually prepared ones in terms of accuracy and completeness,meeting practical requirements.

关 键 词：工作票 电力调度自动化 知识图谱 实体关系抽取 

分 类 号：TP274.2[自动化与计算机技术—检测技术与自动化装置]