基于Transformer的低压停电补全分析  

Complementary Analysis of Low-voltage Power Outage Based on Transformer

作　　者：王大鹏 张永刚 林经伟 钟佳晨 WANG Dapeng;ZHANG Yonggang;LIN Jingwei;ZHONG Jiachen(State Grid Inner Mongolia East Power Co.,Ltd.,Hohhot 010020,China;College of Artificial Intelligence,Nanjing Agricultural University,Nanjing 210095,China)

机构地区：[1]国网内蒙古东部电力有限公司,内蒙古呼和浩特010020 [2]南京农业大学,人工智能学院,江苏南京210095

出　　处：《微型电脑应用》2024年第4期221-225,共5页Microcomputer Applications

摘　　要：相比于建设较为完备的高中压配电网,低压配电网具有分布广、设备运行环境恶劣、监测装置质量参差不齐等特点,使得运行过程中常出现停电事件的漏报、误报、频繁上报等问题,不断增加能源消耗成本并降低电力服务质量。依托智能电表网络架构,最终设计出一种基于深度学习的低压停电补全分析方法。总结了低压配电网的4类停电故障事件,提出深度学习模型Transformer的搭建和使用,设计出与其相关的分析系统架构,并给出可视化管理系统的使用案例。对于低压配电网的停电补全等问题,该方法为业内提供了值得参考的完整解决方案。Compared with the relatively complete high and medium voltage distribution networks,the low-voltage distribution network has the characteristics of wide distribution,harsh equipment operating environment,and uneven quality of monitoring devices,which makes the missed alarm,false alarm and frequent power outage events often occur in the operation process.Problems such as reporting,false reporting,increase the cost of energy consumption and reduce the quality of electricity service.Relying on the smart meter network architecture,a deep learning-based low-voltage power outage completion analysis method is finally designed.The 4 types of power failure events of low-voltage distribution network are summarized,and the construction and use of the deep neural network model transformer is proposed.The related analysis system architecture is designed,and the use case of the visual management system is given.For problems such as power outage completion of low-voltage distribution network,it provides a complete solution worthy of reference for the industry.

关 键 词：低压配电网 低压停电事件补全 深度学习模型 智能管理系统 

分 类 号：TN91[电子电信—通信与信息系统]