基于Sentence-MacBERT模型的同源录波数据匹配方法
CSTR:
作者:
作者单位:

1.河北省分布式储能与微网重点实验室(华北电力大学),河北 保定 071003; 2.国网江西省电力有限公司电力科学研究院,江西 南昌 330096

作者简介:

通讯作者:

中图分类号:

基金项目:

国家自然科学基金项目资助(51877084)


Homologous recording data matching method based on the Sentence-MacBERT model
Author:
Affiliation:

1. Hebei Key Laboratory of Distributed Energy Storage and Microgrid (North China Electric Power University), Baoding 071003, China; 2. Electric Power Research Institute of State Grid Jiangxi Electric Power Co., Ltd., Nanchang 330096, China

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:

    由于不同时期的录波数据记录标准有所不同,以及各个生产厂家对标准的解读存在偏差,造成同源录波数据的通道名称存在个性化差异,且通道索引号不同,难以进行录波数据的同源匹配。针对上述问题,提出基于句向量掩码纠错双向编码器表征语言模型(sentence-masked language model as correction bidirectional encoder representations from transformers, Sentence-MacBERT)的同源录波数据匹配方法。首先,分析录波文件的记录格式特点,根据录波文件的格式特点完成核查信息表的构建。然后,通过构建的核查信息表进行录波文件自动校核。最后,在双向编码器表征(bidirectional encoder representations from transformers, BERT)模型的基础上构建Sentence-MacBERT同源通道匹配模型,完成同源录波数据匹配。算例分析表明,根据核查信息表能够完成录波文件的自动校核,并对解析失败的录波文件发出告警信息。利用Sentence-MacBERT模型进行通道名称匹配的效果良好,能够有效地完成录波数据的同源匹配,帮助运行人员进行故障分析。

    Abstract:

    Because of differences in recording standard over different periods and variations in manufacturers’ interpretation of these standards, homologous recording data often exhibit personalized differences in channel names and channel index numbers, making it difficult to achieve accurate matching of homologous recording data. To solve this problem, a method for matching homologous recording data based on the Sentence-MacBERT model is proposed. First, the characteristics of the recording format are analyzed, and a verification information table is constructed based on these format characteristics. Then, the verification information table is used to automatically verify the recording files. Finally, a Sentence-MacBERT homologous channel matching model is constructed based on the BERT model, and the homologous recording data matching is completed. Case studies show that the verification information table can be used to automatically verify the recording files, and alerts are generated for the recording files that fail to parse. The Sentence-MacBERT model is excellent in channel name matching, effectively completing the homologous matching of recording data and helping operators in analyzing faults.

    参考文献
    相似文献
    引证文献
引用本文

戴志辉,张富泽,韩 笑,等.基于Sentence-MacBERT模型的同源录波数据匹配方法[J].电力系统保护与控制,2025,53(8):159-167.[DAI Zhihui, ZHANG Fuze, HAN Xiao, et al. Homologous recording data matching method based on the Sentence-MacBERT model[J]. Power System Protection and Control,2025,V53(8):159-167]

复制
分享
相关视频

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:2024-07-02
  • 最后修改日期:2024-09-07
  • 录用日期:
  • 在线发布日期: 2025-04-16
  • 出版日期:
文章二维码
关闭
关闭