福州大学学报（自然科学版）2025，Vol.53Issue(2)：135-143,9.DOI:10.7631/issn.1000-2243.24132
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
摘要
Abstract
Addressing the automatic generation of work tickets for power dispatch automation opera-tions,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 algo-rithm 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(F1 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Key words
work order/power dispatch automation/knowledge graph/joint extraction of entity relationships分类
信息技术与安全科学引用本文复制引用
李泽科,陈书里,陈斌,陈荃韡,杨旭,许贻杰,刘延华..基于知识图谱的电力调度自动化作业工作票生成方法[J].福州大学学报（自然科学版）,2025,53(2):135-143,9.基金项目
福建省自然科学基金资助项目(2021J01625) （2021J01625）
国网福建省电力有限公司科技项目(521300230007) （521300230007）
