面向变压器缺陷知识图谱构建的ET-FSUIE关系抽取模型

谢国波; 刘汉林; 林志毅; 叶振豪; 李斯

doi:10.12052/gdutxb.240139

欢迎来到能源电力期刊网！

English Version 期刊登录

您当前的位置：

首页 >

文章列表页 >

面向变压器缺陷知识图谱构建的ET-FSUIE关系抽取模型

电气工程 | 更新时间：2026-05-27

- 面向变压器缺陷知识图谱构建的ET-FSUIE关系抽取模型
- Research on the ET-FSUIE Relation Extraction Model for Transformer Defect Graph Construction
- 广东工业大学学报 2025年42卷第5期页码：77-85
- 作者机构：
  
  1. 广东工业大学计算机学院,广东,广州,510006
  2. 深圳供电局有限公司,广东,深圳,518000
- 作者简介：
  
  谢国波(1977–)，男，教授，博士，主要研究方向为计算机视觉、疾病模式挖掘、电力系统运维等，E-mail：xiegb@gdut.edu.cn
  刘汉林(1998–)，男，硕士研究生，主要研究方向为电力系统运维，E-mail：18087433177@163.com
- 基金信息：
  
  广东省自然科学基金资助项目(2022A1515012379)
- DOI：10.12052/gdutxb.240139
  中图分类号： TM407
- 纸质出版：2025
- 稿件说明：
移动端阅览
谢国波, 刘汉林, 林志毅, 等. 面向变压器缺陷知识图谱构建的ET-FSUIE关系抽取模型[J]. 广东工业大学学报, 2025, 42(5): 77–85.

Xie Guobo, Liu Hanlin, Lin Zhiyi, et al. Research on the ET-FSUIE Relation Extraction Model for Transformer Defect Graph Construction[J]. 2025, 42(5): 77-85.
谢国波, 刘汉林, 林志毅, 等. 面向变压器缺陷知识图谱构建的ET-FSUIE关系抽取模型[J]. 广东工业大学学报, 2025, 42(5): 77–85. DOI： 10.12052/gdutxb.240139

Xie Guobo, Liu Hanlin, Lin Zhiyi, et al. Research on the ET-FSUIE Relation Extraction Model for Transformer Defect Graph Construction[J]. 2025, 42(5): 77-85. DOI： 10.12052/gdutxb.240139.

摘要

在电网数字化运维中，变压器运维数据激增且多为非结构化。这导致缺陷信息提取与故障溯源困难，难以满足智能化运维需求。知识图谱技术为解决这些问题带来了希望，其结构化特性能够整合运维信息，从而提升运维效率。为此，本文提出统一电气设备缺陷短文本格式，构建高质量变压器缺陷关系数据集。然后提出变压器缺陷关系模糊跨度通用信息提取模型(Electrical Transformer-Fuzzy Span Universal Information Extraction

ET-FSUIE)，融合剪枝20%的Roformer(Rotary Transformer) v2预训练语言模型，利用其旋转位置编码处理缺陷描述文本长度差异，提升文本理解能力，并基于 Wasserstein距离改进模糊跨度损失函数 (Wasserstein-Fuzzy Span Loss

W-FSL)，克服传统损失函数局限，提高模型抽取精度。在公开及自建数据集的实验显示，ET-FSUIE模型抽取效果优越，

(F1-Score) 达到81.84%和88.67%。最后，基于该模型抽取的三元组构建电力变压器缺陷关系知识图谱，为电力设备智能化运维提供有力支持。

Abstract

In the context of digital operation and maintenance of power grids

the surge in unstructured transformer data has made defect information extraction and fault tracing challenging

hindering the transition to intelligent maintenance. Knowledge graph technology offers a potential solution by leveraging its structured nature to integrate operational data and improve efficiency. Inspired by this

this paper proposes a unified short-text format for electrical equipment

defects and constructs a high-quality transformer defect relationship dataset. The ET-FSUIE (Electrical Transformer-Fuzzy Span Universal Information Extraction) model is introduced

which integrates a 20% pruned Roformer v2 pre-trained language model. By utilizing its rotary position encoding

the model effectively handles variations in defect description text lengths

enhancing text comprehension. Additionally

a W-FSL (Wasserstein-Fuzzy Span Loss) loss function based on Wasserstein distance is proposed to overcome the limitations of traditional loss functions and improve extraction accuracy. Experimental results on both public and self-built datasets demonstrate the superior performance of the ET-FSUIE model

achieving

scores of 81.84% and 88.67%. Finally

a knowledge graph for power transformer defect relationships is constructed using the extracted triplets

providing robust support for the intelligent transformation of power equipment operation and maintenance.

关键词

Keywords

references

LI X B, TIAN F Q, GUO Y K. Key technologies for health management and intelligent operation and maintenance of power equipment in new power systems[J]. Power System Technology, 2023, 47(9): 3710-3726.

ZHU Y M, XU X Y, YAN Z, et al . Prospect of renewable energy integrated distribution network operation in the power internet of things[J ] . Power System Protection and Control, 2022, 50(2): 176-187.

ZHAO P, PU T J, WANG X Y, et al . Key technologies and p erspectives of power internet of things facing with digital twins of the energy internet[J ] . Proceedings of the CSEE. 2022, 42(2): 447-457.

ZHANG G Z, CHEN K, FANG R X, et al . Transformer fault diagnosis based on DGA and a whale algorithmoptimizing a LogitBoost-decision tree[J ] . Power System Protection and Control, 2023, 51(7): 63-72.

XIANG C M, ZENG S M, YAN P, et al . Typical application and prospect of digital twin technology in power grid operation[J ] . High Voltage Engineering, 2021, 47(5): 1564-1575.

LI G, LI Y Q, WANG H T, et al . Knowledge graph of power equipment health management: basic concepts, key technologies and research progress[J ] . Automation of Electric Power Systems, 2022, 46(3): 1-13.

迟人俊. 基于深度神经网络的中文实体关系抽取研究与实现[D]. 北京: 北京邮电大学, 2020.

CHEN Z, GUO C. A pattern-first pipeline approach for entity and relation extraction[J]. Neurocomputing, 2022, 494: 182-191.

LIU X, SONG Q, ZHANG P Z. Relation extraction based on deep learning[C]//2018 IEEE/ACIS 17th International Conference on Computer and Information Science. Singapore: IEEE, 2018: 687-691.

GENG Z, ZHANG Y, HAN Y. Joint entity and relation extraction model based on rich semantics[J]. Neurocomputing, 2021, 429: 132-140.

LOU J, LUY J, DAI D, et al . Universal information extraction as unified semantic matching[C ] //Proceedings of the AAAI Conference on Artificial Intelligence. Washington DC: AAAI, 2023: 37(11): 13318-13326.

XU D, CHEN W, PENG W, et al . Large language models for generative information extraction: a survey[J ] . Frontiers of Computer Science, 2024, 18(6): 186357.

SHU J W, YANG T, GENG Y N, et al . A joint extraction method for overlapping entity relationships in the construction of electric power knowledge graph[J ] . High Voltage Engineering, 2024, 50(11): 4912-4922.

XIE Q, CAI Y, XIE J, et al . Research on construction method and application of knowledge graph for power transformer operation and maintenance based on ALBERT[J ] . Transactions of China Electrotechnical Society, 2023, 38(1): 95-106.

SU J L, AHMED M, LU Y, et al . Roformer: enhanced transformer with rotary position e mbedding[J ] . Neurocomputing, 2024, 568: 127063.

PANARETOS V M, ZEMEL Y. Statistical aspects of wasserstein distances[J]. Annual Review of Statistics and Its Application, 2019, 6(1): 405-431.

LI S, HE W, ShI Y, et al . Duie: a large-scale chinese dataset for information extraction[C ] //Natural Language Processing and Chinese Computing: 8th CCF International Conference. Dunhuang: NLPCC, 2019: 791-800.

SHANG Y M, HUANG H, MAO X. Onerel: joint entity and relation extraction with one module in one step[C]//Proceedings of the AAAI conference on artificial intelligence. Oxford United Kingdom: AAAI, 2022, : 36(10): 11285-11293.

CUI Y, CHE W, LIU T, et al . Pre-training with whole word masking for chinese bert[J ] . IEEE/ACM Transactions on Audio, Speech, and Language Processing, 2021, 29: 3504-3514.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

概率条件下基于双目标交替优化的知识表示模型

知识图谱的拓展及其智能拓展库

基于路况参数的混合动力机车能量管理策略研究

融合沟道倾斜角的3D NAND存储单元建模及编程策略优化

基于数字混频的脉冲式激光测距方法的研究与实现