基于多源数据融合的电网调控行为异常监测与诊断
CSTR:
作者:
作者单位:

(1.广东电网有限责任公司电力调度控制中心,广州 510220;2.南京南瑞信息通信科技有限公司,南京 211100)

作者简介:

吴岳洲(1971-),男,广东省广州市人,高级工程师,主要研究方向为电力系统。

通讯作者:

中图分类号:

TN911. 7;TP274

基金项目:


Multi-Source Data-Fusion-Mediated Abnormality Monitoring and Diagnosis of Power Grid Regulation Behavior
Author:
Affiliation:

(1. Power Dispatching and Controlling Center of Guangdong Power Grid Co.,Ltd.,Guangzhou 510220 CHN;2. Nanjing NARI Information and Communication Technology Co.,Ltd.,Nanjing 211100,CHN)

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:

    在主电网断路器操动机构异常监测中,多维混合特征提取困难,导致分类识别准确率较低。特别是当电网处于恶劣场景中,信号干扰和数据不完整的问题更加突出,进一步增加监测和故障诊断的难度。对此,文章提出一种基于变分模态分解算法与综合式梅尔倒谱系数的多源信号特征提取方法。在该方法的基础上,利用核主成分聚类分析与优化后的支持向量机,最终提出一种新型的主电网断路器操动机构异常监测与诊断技术。实验结果表明,在正常工况下,该研究方法信号的特征提取标准差与均方根值分别为 0. 10与 1. 12,验证了其在多源信号特征提取与噪声抑制中的有效性。此外,该方法的最高分类识别准确率达到了 97. 7%,显示出其在提高异常监测精度和诊断效率方面具有显著优势。所提出的方法不仅能为主电网断路器操动机构的异常监测与诊断提供新的技术路径,还能为实际应用中的多源信号分析提供有效的解决方案,解决军事电网中信号干扰和数据不完整的问题,提升监测与诊断的智能化水平。

    Abstract:

    Monitoring the abnormalities in the operating mechanism of the main-grid circuit breaker through multi-dimensional hybrid feature extraction is challenging owing to its poor classification performance and low identification accuracy. Therefore,in this study,a multi-source signal-feature extraction technique, based on a variational modal decomposition algorithm with integrated Mayer inverse spectral coefficients,is proposed. This technique is integrated with kernel principal component cluster analysis and optimized support vector machine to develop a new type of method for monitoring and diagnosing the main power-grid circuit breaker operating mechanism. The experimental results show that under normal operating conditions,the standard deviation and root-mean-square value of the proposed signal-feature extraction method are 0. 10 and 1. 12,respectively, which verify the effectiveness of the proposed method in multi-source signal-feature extraction and noise suppression. In addition,the proposed method achieves the highest classification recognition accuracy of 97. 7%. The results demonstrate that the proposed method can improve the accuracy of anomaly monitoring and diagnostic efficiency. It provides a novel technical path for anomaly monitoring and diagnosis of the operating mechanism of the main grid circuit breaker. In addition,the proposed method offers an effective solution for the analysis of multi-source signals in practical applications.

    参考文献
    相似文献
    引证文献
引用本文

吴岳洲,蓝天,陈志伟,缪新新,王振宇.基于多源数据融合的电网调控行为异常监测与诊断[J].半导体光电,2026,47(1):140-148. WU Yuezhou, LAN Tian, CHEN Zhiwei, MIAO Xinxin, WANG Zhenyu. Multi-Source Data-Fusion-Mediated Abnormality Monitoring and Diagnosis of Power Grid Regulation Behavior[J].,2026,47(1):140-148.

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:2025-08-14
  • 最后修改日期:
  • 录用日期:
  • 在线发布日期: 2026-03-20
  • 出版日期:
文章二维码