摘要
提出一种针对配电网线损异常数据问题的多级辨识与修正方法。首先,运用基于DBSCAN-新息序列算法做初级辨识,识别出异常数据和可疑数据,进而根据线损数据的时间惯性对可疑数据进行二次辨识以减少误判率;然后,采用改进的LSTM算法对异常数据进行修正;最后,在IEEE-69节点配电系统中应用甘肃临夏某配电台区的实际线损数据验证了所提方法的有效性。
This paper presents a multi-level identification and correction method to address the problem of abnormal line loss data in distribution networks.Firstly, the primary identification based on density-based spatial clustering of applications with noise(DBSCAN)clustering algorithm and innovation sequence algorithm is used to identify abnormal data and suspicious data.Then, according to the time inertia of the line loss data, the suspicious data is identified again to reduce the misjudgment rate.In addition, the improved long short term memory(LSTM)algorithm is used to correct the abnormal data.Finally, the effectiveness of the proposed method is verified by the actual line loss data of a distribution station in Linxia, Gansu Province in the IEEE-69 node distribution system.
作者
夏懿
丁坤
马慧莲
王鹏
张铄
XIA Yi;DING Kun;MA Huilian;WANG Peng;ZHANG Shuo(Linxia Power Supply Company,State Grid Gansu Electric Power Company,Linxia 731100,China;College of Electrical and Information Engineering,Lanzhou University of Technology,Lanzhou 730050,China)
出处
《机械与电子》
2023年第2期13-17,共5页
Machinery & Electronics
基金
国网甘肃省电力公司科技项目(SGGSLX00FCJS2200358)。
关键词
线损数据
多级辨识
异常数据
改进LSTM算法
line loss data
multistage identification
abnormal data
improved LSTM algorithm
作者简介
夏懿(1974-),男,山东聊城人,学士,高级工程师,研究方向为电网信息化研究和管理。
