摘要
为了提升低压配电网台区故障区段精准定位效果,该文提出基于多模态数据感知的低压配电网台区故障区段精准定位方法。该方法运用主成分分析法确定多模态暂态零序电流数据的综合指标,并采集相关数据。通过相似性指标提取特征参数,并利用循环神经网络自适应融合各监测点的数据。计算监测点间暂态零序电流的相关系数,并与预设阈值比较,以判定故障区段。实验结果表明,该方法能有效感知并融合多模态数据,且在不同故障角下均能准确地定位故障区段。
In order to improve the precise positioning effect of fault sections in low⁃voltage distribution network substations,a research method based on multi⁃modal data perception for precise positioning of fault sections in low⁃voltage distribution network substations is proposed.The principal component analysis method is used to determine the comprehensive indicators of multi⁃modal transient zero sequence current data,and relevant data is collected.Feature parameters are extracted through similarity indicators,and a recurrent neural network is used to adaptively fuse data from various monitoring points.Calculate the correlation coefficient of transient zero sequence current between monitoring points and compare it with the preset threshold to determine the fault section.The experimental results show that this method can effectively perceive and fuse mulri⁃modal data,and accurately locate fault segments at different fault angles.
作者
吴迪
李强
毕坤
龙丹
程时润
WU Di;LI Qiang;BI Kun;LONG Dan;CHENG Shirun(State Grid Ezhou Power Supply Company,Ezhou 436000,China)
出处
《电子设计工程》
2024年第17期122-126,共5页
Electronic Design Engineering
关键词
多模态数据感知
低压配电网
故障区段
精准定位
主成分分析
相关系数
multi⁃modal data sensing
low⁃voltage distribution network
fault section
precise positioning
principal component analysis
correlation coefficient
作者简介
吴迪(1981-),男,湖北鄂州人,高级工程师。研究方向:电网数字化管理。
