基于边缘计算的台区用户用电特性感知方法  

Edge computing based user power characteristics sensing method in platform

作　　者：李金灿 李景顺 覃江英[3] LI Jincan;LI Jingshun;QIN Jiangying(Guangxi Power Grid Co.,Ltd.,Nanning 530000,China;Guilin Power Supply Bureau of Guangxi Power Grid Corporation,Guilin 541000,China;Hechi Power Supply Bureau of Guangxi Power Grid Corporation,Hechi 547000,China)

机构地区：[1]广西电网有限责任公司,广西南宁530000 [2]广西电网公司桂林供电局,广西桂林541000 [3]广西电网公司河池供电局,广西河池547000

出　　处：《电子设计工程》2024年第3期50-53,58,共5页Electronic Design Engineering

基　　金：广西电网公司科技项目(GXKJXM20200427)。

摘　　要：目前不同地区、不同时间的负荷不能直接叠加,导致台区用户属性判定结果不精准,感知效果较差。为解决这一问题,提出基于边缘计算的台区用户用电特性感知方法。使用双向台区用户识别仪器识别相邻台区干扰严重的载波信号。对台区用电风险大数据进行均衡调度处理,结合边缘计算构建属性判定函数。根据用户属性判定结果,设计用电特性感知流程,计算平均采集成功率,提取台区用户用电特性,生成异常用户告警清单。对原始日负荷曲线进行标么化处理,判断其是否满足聚类误差指标,去掉疑似异常用户。计算台区线损百分比,感知高压台区和低压台区。实验结果表明,该方法第一、二类用户在70%尖峰负荷下的持续时间均比实际数据高20 h,具有精准感知效果。Currently,loads in different regions and at different times cannot be directly superimposed,resulting in inaccurate user attribute determination results in station areas and poor perception effects.In order to solve this problem,an edge computing⁃based perception method of users’power consumption characteristics in station areas is proposed.Use the two⁃way station area user identification instrument to identify the carrier signal with serious interference in the adjacent station area.Balanced scheduling is performed on the big data of electricity consumption risks in the station area,and an attribute determination function is constructed in combination with edge computing.According to the user attribute determination results,the power consumption characteristic perception process is designed,the average acquisition success rate is calculated,the power consumption characteristics of users in the station area are extracted,and an alarm list of abnormal users is generated.The original daily load curve is subjected to standardization processing to determine whether it meets the clustering error index,and the suspected abnormal users are removed.Calculate the percentage of line loss in the station area,and perceive the high⁃pressure station area and the low⁃pressure station area.The experimental results show that the duration of the first and second types of users under the 70%peak load is 20 h longer than the actual data,which has the accurate perception effect.

关 键 词：边缘计算 台区用户 用电特性 感知 

分 类 号：TP202[自动化与计算机技术—检测技术与自动化装置]