基于台区用电功率跃变特征序列的拓扑识别方法

Topology Identification Method Based on Features Sequence of Electric Power Consumption in Low-voltage Areas

针对低压配电网台区长期存在的拓扑结构不清晰、更新滞后,人力维护工作量大等问题,提出了一种基于台区用电功率跃变特征序列的自动拓扑识别方法。该方法使用具备一定边缘计算能力的变压器智能采集终端与表箱智能采集终端采集分析台区用电特征信号序列,通过网络上送主站数据平台,再由主站数据平台做进一步分析运算,得出各台区下电表与变压器的从属关系,确定台区拓扑。根据试点小区实验效果,采用所述的拓扑识别系统,选取吻合度值5作为判断阈值,拓扑正确率能达99%以上,能够很好地划分出台区户变从属拓扑关系。该方法结合恰当的匹配阈值表现出较显著的拓扑识别成效,并降低了人力维护成本,具备在实际工程中推广应用的价值。

In response to the long-standing issues in low-voltage distribution substations,such as unclear topolo⁃gy structure,outdated updates,and high workload for manual maintenance,an automatic topology identifica⁃tion method based on the power consumption transient feature sequence of substations has been proposed.This method uses transformer data terminal units(TDTUs)and meter data terminal units(MDTUs)which have edge computing capabilities to collect and analyze the power consumption feature signal sequences of substa⁃tions.The data is then transmitted to the main station data platform via the network,where further analysis and calculation are carried out to determine the relationship between electric meters and transformers in each substa⁃tion,thus establishing the topology of substations.According to the pilot experiment results in selected residen⁃tial areas,the topology identification system performs well with a chosen threshold value of 5,achieving a topol⁃ogy accuracy rate of over 99%,and effectively determining the relationship between households and transform⁃ers in substations.This method,combined with an appropriate matching threshold,demonstrates significant topology identification results and greatly reduces the cost of manual maintenance,making it valuable for practical engineering applications and promotion.