摘要
传统变压器异常状态检测算法的有效性与精度仍然有待进一步研究。文章提出了采用投影滑动窗口和可拓K-means聚类相结合的变压器状态异常检测一般模型及分析方法。将空间状态变量投影到坐标轴加以考虑,在可拓距离度量下构造指标关联函数,按照给定的规则建立异常检测模型。以湖州某变电站各项历史监测数据对上述模型进行算例分析,结果表明,该方法可定量分析在线监测数据与变压器异常状态之间的动态关系,为实施变压器状态异常检测提供了一定的理论依据。
The validity and accuracy of traditional transformer abnormal state detection still need to be further studied.Therefore,a power transformer anomaly detection model and analysis method using sliding window projection and extension K-means clustering was proposed.Spatial state variables were projected to the coordinate axis,the index correlation function was constructed in the extension distance metric,and the anomaly detection model was established according to the given rules.The model was verified by the case analysis based on the monitoring data from a substation in Huzhou.The results show that the method can quantitatively analyze the dynamic relationship between the abnormal state of the transformer and the online monitoring data,which provides a theoretical basis for the implementation of anomaly detection for transformers.
作者
林越
刘廷章
曾福庚
LIN Yue;LIU Tingzhang;ZENG Fugeng(School of Ocean Information Engineering,Hainan Tropical Ocean University,Sanya 572022,China;School of Mechatronic Engineering and Automation,Shanghai University,Shanghai 200072,China)
出处
《合肥工业大学学报（自然科学版）》
CAS
北大核心
2019年第9期1175-1180,1185,共7页
Journal of Hefei University of Technology：Natural Science
基金
国家自然科学基金资助项目(61273190)
海南省自然科学基金资助项目(20166223)
海南热带海洋学院2018年校级青年专项基金资助项目(RHDQN201818)
关键词
可拓理论
聚类算法
变压器
异常检测
extension theory
clustering algorithm
power transformer
abnormal state detection
