基于VMD-DE-ELM的同杆双回输电线路故障识别  

Fault Identification of Double-Circuit Transmission Line on the Same Pole Based on VMD-DE-ELM

作　　者：邹西 吴浩 邓思敬 漆知渊 宋弘 ZOU Xi;WU Hao;DENG Sijing;QI Zhiyuan;SONG Hong(School of Automation and Information Engineering,Sichuan University of Science&Engineering,Yibin 644000,China;Artificial Intelligence Key Laboratory of Sichuan Province,Yibin 644000,China;College of Electronic Information and Automation,Aba Teachers University,Aba Prefecture 623002,China)

机构地区：[1]四川轻化工大学自动化与信息工程学院,四川宜宾644000 [2]人工智能四川省重点实验室,四川宜宾644000 [3]阿坝师范学院电子信息与自动化学院,四川阿坝州623002

出　　处：《四川轻化工大学学报（自然科学版）》2023年第5期41-50,共10页Journal of Sichuan University of Science & Engineering(Natural Science Edition)

基　　金：四川省科技厅项目(2020YFG0178,2021YFG0313,2022YFS0518,2022ZHCG0035);人工智能四川省重点实验室项目(2019RYY01);四川理工学院四川省院士(专家)工作站项目(2018YSGZZ04)。

摘　　要：为了提升同杆双回输电线路的稳定性和准确性,通过对区内/外故障电压反行波变化规律进行分析,提出了一种基于变分模态分解和差分进化算法优化极限学习机(VMD-DE-ELM)的同杆双回输电线路区内/外故障识别新方法。首先对发生故障后两端的电压、电流进行相模变换;再利用VMD将故障后一段时窗内的电压反行波分解到5个尺度上;用特征提取对应尺度下的能量熵组成特征向量;最后针对区内/外故障样本具有不平衡性,通过使用SMOTE算法对区外样本进行扩充后,将特征向量集输入到DE-ELM分类器进行训练和测试。大量仿真结果表明:该方法在不同故障类型、不同过渡电阻、不同故障初始角以及不同故障位置情况下能有效实现区内外故障识别,且在CT饱和、噪声干扰等情况下也能较好识别区内外故障。In order to improve the stability and accuracy of the double-circuit transmission line on the same pole,a new internal/external faults identification method of the double-circuit transmission line on the same pole based on variational mode decomposition and differential evolution&extreme learning machine(VMD-DE-ELM)has been proposed by analyzing the variation law of internal/external fault voltage anti-travelling wave.Firstly,the phase-mode of the voltage and current at both ends are transformed after the fault occurs;then the VMD is used to decompose the voltage anti-travelling wave in a period of time after the fault into five scales,whose energy entropy obtained by feature extraction is used to form the feature vector;aiming at the imbalance of the internal/external fault samples,the feature vector set is finally input into the DE-ELM classifier for training and testing after expanding the out-of-zone samples by using the SMOTE algorithm.A large number of simulation results show that the method can effectively realize the fault identification inside and outside the area under the conditions of different fault types,different transition resistances,different initial fault angles and different fault locations,and it is also easy to identify faults inside and outside the zone under the conditions of CT saturation,noise interference.

关 键 词：同杆双回 电压反行波 变分模态分解 SMOTE算法 差分进化算法优化极限学习机 故障识别 

分 类 号：TM75[电气工程—电力系统及自动化]