摘要
低压配电台区是窃电行为的主要发生区域之一,分布式光伏接入使得台区由传统单电源结构变为多电源结构,线损特征改变,窃电检测难度进一步提升。面向低压光伏台区,提出了一种基于多维序列分析的窃电检测方法。首先提取多维线损率影响因素,基于高斯过程回归进行非线性拟合并通过交叉验证确定窃电嫌疑时段;然后借助电量、电压等多维序列信息计算各用户的理论线损贡献因子,结合相关性分析和蒙特卡洛模拟捕捉电量线损耦合关系;进而基于OPTICS聚类定位嫌疑度离群用户,即窃电用户;最后通过仿真台区算例分析验证了所提方法的有效性。
Low-voltage transformer district is one of the main areas where energy theft occurs.The access of distributed photovoltaic makes the structure of the transformer district change from traditional single power supply to multi-power supply,which changes the characteristics of line loss and further increases the difficulty of energy theft detection.A detection method of energy theft based on multi-dimensional series analysis is proposed for low-voltage photovoltaic transformer district.Firstly,multi-dimensional factors affecting line loss rate are extracted,nonlinear fitting is carried out based on Gaussian process regression,and the suspected period of energy theft is determined by cross verification.Secondly,the theoretical line loss contribution factor of each user is calculated by using multi-dimensional series information such as power and voltage,and the coupling relationship between the power and line loss is captured by correlation analysis and Monte Carlo simulation.Then,users with outlier suspicion are identified based on OPTICS clustering,that is,energy theft users.Finally,the effectiveness of the proposed method is verified by the examples in the simulation transformer district.
作者
盛锦昊
张蓬鹤
杨艺宁
赵波
SHENG Jinhao;ZHANG Penghe;YANG Yining;ZHAO Bo(Beijing Information Science and Technology University,Beijing 100192,China;China Electric Power Research Institute,Beijing 100192,China)
出处
《电测与仪表》
北大核心
2024年第9期216-224,共9页
Electrical Measurement & Instrumentation
基金
国家电网有限公司总部科技项目(5700-2022 55222A-1-1-ZN)。
关键词
窃电检测
低压光伏台区
线损率
高斯过程回归
蒙特卡洛模拟
OPTICS聚类
energy theft detection
low-voltage photovoltaic transformer district
line loss rate
Gaussian process regression
Monte Carlo simulation
OPTICS clustering
