周健科,王震,成杰,郑宇鹏,汤波,李铭杰
ZHOU Jianke,WANG Zhen,CHENG Jie,ZHENG Yupeng,TANG Bo,LI Mingjie

• 1:国网浙江省电力有限公司舟山供电公司
• 2:上海电力大学电气工程学院

摘要(Abstract)：

设备线夹起着连接、紧固线路的作用，其健康状况直接影响供电可靠性。针对海洋环境下设备线夹失效造成线路故障占比高的问题，进行设备线夹失效机理分析并提出一种基于A-DBiLSTM（改进神经网络）的寿命评估方法。首先，通过设备线夹运行环境模拟实验建立失效进程与环境因素的关联模型，并提取关键故障特征，制定寿命状态评估指标；其次构建A-DBiLSTM模型，考虑多环境因素、强相关性环境特征对线夹运行状态的影响，输入环境特征和评估指标，并通过Attention机制重点赋权典型强气象特征，实现对线夹的寿命评估；最后基于浙江某海岛配电网设备线夹实际运行数据进行验证，结果表明：该方法能够准确评估配电网设备线夹在不同海洋环境下的寿命状态，对保障线路可靠性、改进线夹运行管理及提升设备质量具有实用价值。
Device clamps are used for connecting and fastening lines, and their health condition directly affects the reliability of power supply. In view of the line faults largely caused by the failure of device clamps under marine environment, the clamp failure mechanism is analyzed and a life assessment method based on an improved neural network A-DBiLSTM is introduced. Firstly, a correlation model between the failure process and environmental factors is established through the simulation experiment of device clamp operation environment, and the key failure characteristics are extracted to develop the life state assessment index. Secondly, the A-DBiLSTM model is constructed, and the influence of multi-environmental factors and highly correlated environmental characteristics on the operating state of clamps is considered. The life assessment of clamps is achieved by focusing on the assignment of typical strong meteorological characteristics through the Attention mechanism. Finally, the method is validated based on the actual operation data of device clamps in an island distribution network in Zhejiang province. The results show that the method can accurately assess the life state of device clamps of distribution networks under different marine environments and is of practical value for ensuring line reliability, improving clamp operation management, and enhancing device quality.

关键词(KeyWords)： 配电网;设备线夹;失效机理;改进神经网络;Attention机制;寿命评估
distribution networks;device clamp;failure mechanism;improved neural network;Attention machnism;life assessment

Abstract:

Keywords:

基金项目(Foundation): 国网浙江省电力有限公司科技项目（5211ZS190071）

作者(Author): 周健科,王震,成杰,郑宇鹏,汤波,李铭杰
ZHOU Jianke,WANG Zhen,CHENG Jie,ZHENG Yupeng,TANG Bo,LI Mingjie

DOI: 10.19585/j.zjdl.202304002

参考文献(References)：

• [1]马钊，周莉梅，袁海文，等.现代配电网健康指数理论及其应用研究[J].中国电机工程学报，2019,39(1):130-140.MA Zhao,ZHOU Limei,YUAN Haiwen,et al.Health index methodology of modern distribution network and its application[J]. Proceedings of the CSEE,2019,39(1):130-140.
• [2] O'DONNELL J,O'DONNELL J E D.Coastal vulnerability in Long Island Sound:the spatial structure of extreme sea level statistics[C]//2012 Oceans. 14-19 Oct. 2012,Hampton Roads,VA,USA:IEEE,2013:1-4.
• [3]吴勇军，薛禹胜，谢云云，等.台风及暴雨对电网故障率的时空影响[J].电力系统自动化，2016,40(2):20-29.WU Yongjun,XUE Yusheng,XIE Yunyun,et al.Spacetime impact of typhoon and rainstorm on power grid fault probability[J]. Automation of Electric Power Systems,2016,40(2):20-29.
• [4] CIGREWG.SCB2.54. Guidelines for the management of risk associated with severe climatic events and climate change on overhead lines[R].Paris,France:CIGRE,2014.
• [5]陈国宏，季坤，严波，等.高压输电耐张线夹的腐蚀行为及其影响因素[J].机械工程材料，2017,41(4):74-79.CHEN Guohong,JI Kun,YAN Bo,et al.Corrosion behavior and influencing factors of strain clamp for high voltage electricity transmission[J].Materials for Mechanical Engineering,2017,41(4):74-79.
• [6]周经中，何学敏，孙阔腾，等.强腐蚀地区输电线路腐蚀及监测防护的研究现状[J].腐蚀与防护，2021,42(4):1-8.ZHOU Jingzhong,HE Xuemin,SUN Kuoteng,et al.Research status of corrosion and monitoring protection of transmission lines in serious corrosion area[J]. Corrosion&Protection,2021,42(4):1-8.
• [7]刘琴，李瑶琴，谢梁，等.盐雾环境下清洁和染污绝缘子交流耐压特性[J].高电压技术，2018,44(9):2828-2834.LIU Qin,LI Yaoqin,XIE Liang,et al.AC withstand characteristics of clean and polluted insulators in salt-fog[J].High Voltage Engineering,2018,44(9):2828-2834.
• [8]电力行业电力规划设计标委会.10 kV及以下架空配电线路设计技术规程：DL/T 5220—2005[S].北京：中国电力出版社，2005.
• [9]汤世彬，赵潇，杨晓龙.500 kV紧凑型杆塔耐张线夹连接板断裂分析[J].浙江电力，2020,39(10):69-73.TANG Shibin,ZHAO Xiao,YANG Xiaolong.Analysis of connecting plate fracture of tension clamp of 500 kV compact towers[J]. Zhejiang Electric Power,2020,39(10):69-73.
• [10]赵洲峰，王若民，陈国宏，等.基于数字射线的耐张线夹压接质量评价试验[J].浙江电力，2018,37(11):41-45.ZHAO Zhoufeng,WANG Ruomin,CHEN Guohong,et al.Quality assessment test of strain camp based on digital radiography[J]. Zhejiang Electric Power,2018,37(11):41-45.
• [11]裘吟君，刘岩，杨大宁，等.沿海城市电网户外用线夹与螺栓耐蚀性研究[J].南方电网技术，2008,2(6):78-80.QIU Yinjun,LIU Yan,YANG Daning,et al.Study on the corrosion resistance of the outdoor insulating clamps and bolts in coastal-city power networks[J]. Southern Power System Technology,2008,2(6):78-80.
• [12]汤晓磊，陈国宏，刘俊建，等.平面接触型铜铝过渡设备线夹的大气腐蚀行为[J].电力建设，2014,35(4):65-69.TANG Xiaolei,CHEN Guohong,LIU Junjian,et al.Atmospheric corrosion behavior of Cu/Al plane contact type terminal connectors[J]. Electric Power Construction,2014,35(4):65-69.
• [13]董瑾，钟相源，安江英.变电站铜铝过渡设备线夹断裂原因[J].中国电力，2010,43(12):27-30.DONG Jin,ZHONG Xiangyuan,AN Jiangying. Fracture of copper and aluminum middle-sized device clamp in substation[J].Electric Power,2010,43(12):27-30.
• [14]余长厅，黎大健，汲胜昌，等.基于振动噪声及BP神经网络的变压器故障诊断方法研究[J].高压电器，2020,56(6):256-261.YU Changting,LI Dajian,JI Shengchang,et al.Research on transformer fault diagnosis method based on vibration noise and BP neural network[J].High Voltage Apparatus,2020,56(6):256-261.
• [15]代杰杰，宋辉，盛戈皞，等.采用LSTM网络的电力变压器运行状态预测方法研究[J].高电压技术，2018,44(4):1099-1106.DAI Jiejie,SONG Hui,SHENG Gehao,et al.Prediction method for power transformer running state based on LSTM network[J]. High Voltage Engineering,2018,44(4):1099-1106.
• [16]杨志淳，靖晓平，乐健，等.基于MI-PSO-BP算法的配电设备状态实时评估方法[J].电力自动化设备，2019,39(12):108-113.YANG Zhichun,JING Xiaoping,LE Jian,et al.Real-time condition assessment method based on MI-PSO-BP algorithm for distribution equipment[J].Electric Power Automation Equipment,2019,39(12):108-113.
• [17]国家质量监督检验检疫总局，中国国家标准化管理委员会.电工电子产品环境试验第2部分：试验方法试验Db：交变湿热（12 h+12 h循环）：GB/T 2423.4—2008[S].北京：中国标准出版社，2009.
• [18]国家质量监督检验检疫总局，中国国家标准化管理委员会.人造气氛腐蚀试验盐雾试验：GB/T 10125—2012[S].北京：中国标准出版社，2013.
• [19]国家质量监督检验检疫总局，中国国家标准化管理委员会.电工电子产品环境试验第2部分：试验方法试验：温度（低温、高温）/低气压/振动（正弦）综合：GB/T2423.102—2008[S].北京：中国标准出版社，2008.
• [20]帅勇，胡峰，朱悌峰.常德电网变电设备导电回路过热缺陷原因及对策分析[J].高压电器，2014,50(5):120-125.SHUAI Yong,HU Feng,ZHU Tifeng.Analysis on overheat of conductive circuit of transformer equipment in Changde power grid with countermeasures[J].High Voltage Apparatus,2014,50(5):120-125.
• [21]吴晓欣，何怡刚，段嘉珺，等.考虑复杂时序关联特性的Bi-LSTM变压器DGA故障诊断方法[J].电力自动化设备，2020,40(8):184-193.WU Xiaoxin,HE Yigang,DUAN Jiajun,et al.Bi-LSTMbased transformer fault diagnosis method based on DGA considering complex correlation characteristics of time sequence[J].Electric Power Automation Equipment,2020,40(8):184-193.
• [22]何健明，李梦诗，张禄亮，等.基于Attention和残差网络的非侵入式负荷监测[J/OL].电测与仪表：1-9[2021-08-13]. http：//kns. cnki. net/kcms/detail/23.1202.TH.20210806.1314.002.html.HE Jianming,LI Mengshi,ZHANG Luliang,et al. Nonintrusive load monitoring algorithm based on attention and residual networks[J/OL]. Electrical Measurement&Instrumentation:1-9[2021-08-13]. http：//kns. cnki. net/kcms/detail/23.1202.TH.20210806.1314.002.html.
• [23]李玉志，刘晓亮，邢方方，等.基于Bi-LSTM和特征关联性分析的日尖峰负荷预测[J].电网技术，2021,45(7):2719-2730.LI Yuzhi,LIU Xiaoliang,XING Fangfang,et al. Daily peak load prediction based on correlation analysis and Bidirectional long short-term memory network[J]. Power System Technology,2021,45(7):2719-2730.
• [24]赵文清，沈哲吉，李刚.基于深度学习的用户异常用电模式检测[J].电力自动化设备，2018,38(9):34-38.ZHAO Wenqing,SHEN zheji,LI Gang. Anomaly detection for power consumption pattern based on deep learning[J].Electric Power Automation Equipment,2018,38(9):34-38.
• [25]周秀，朱洪波，马云龙，等.基于深度学习的变压器局部放电模式识别研究[J].高压电器，2019,55(12):98-105.ZHOU Xiu,ZHU Hongbo,MA Yunlong,et al.Partial discharge pattern recognition of transformer based on deep learning[J]. High Voltage Apparatus,2019,55(12):98-105.